fix(agents,hands): per-agent/per-hand mcp_servers / skills allowlists (#87) (#92)

All 32 agent manifests and 17 hands shipped with empty mcp_servers /
skills lists, which the kernel interprets as "no filter" — every
globally-configured MCP server's tools and every installed skill get
injected into the prompt on every LLM call. On a typical instance (9
MCP servers, ~85 MCP tools + ~82 built-in tools) that's ~50k input
tokens per turn spent on definitions the agent never uses.

Changes
-------

32 agents/*/agent.toml:
  - mcp_servers: 1-4 per agent. memory wherever state persists across
    turns; fetch / exa-search / brave-search only where the prompt
    actually calls for web; git / github / filesystem on engineering
    agents; gmail / google-calendar / linear / jira on productivity
    agents whose prompts mention them.
  - skills: per-role allowlist driven by what the system_prompt names
    (e.g. coder → rust/python/typescript/git/shell-scripting; devops-
    lead → docker/kubernetes/terraform/ansible/ci-cd/helm/prometheus/
    sysadmin). Generalists (assistant) keep skills = [] (see "Open
    items" below).
  - skills_disabled = true on the four short-conversational agents
    (hello-world, recipe-assistant, health-tracker, home-automation).
    Their system prompts never instruct the LLM to consult any skill,
    so loading all 60 was pure waste. They also drop the explicit
    max_history_messages override and inherit the kernel default (60).
  - max_history_messages tiered by workload shape:
      60  short conversational (hello-world, recipe, health-tracker,
          home-automation) — inherits the rising kernel default
          (`DEFAULT_MAX_HISTORY_MESSAGES = 60`); no override needed.
      60  single-turn task agents (writer, translator, doc-writer,
          email-assistant, customer-support, sales-assistant, recruit-
          er, social-media, personal-finance, tutor, travel-planner,
          meeting-assistant, ops, devops-lead, planner) — explicit
          override at the same value to lock the cap if the kernel
          default moves again.
      80  multi-step / tool-heavy (coder, debugger, architect, code-
          reviewer, test-engineer, security-auditor, analyst, data-
          scientist, academic-researcher, researcher, legal-assistant)
      120 coordinators (assistant, orchestrator) — long multi-agent
          sessions where prompt-cache continuity is critical
    All values sit at or above the kernel default. Pinning lower
    would thrash the prompt cache (the failure mode #91 fixed for
    the creator hand by *raising* the cap, not lowering it).

17 hands/*/HAND.toml:
  - hand-level mcp_servers / skills now declared on every hand, so
    every [agents.*] inside inherits a sensible allowlist.
  - skills_disabled = true placed on each [agents.*] inside clip and
    creator (pure media pipelines that don't benefit from any skill).
    HandDefinitionRaw in librefang-hands does NOT have a top-level
    skills_disabled field — declaring it at the hand top level would
    be silently dropped by serde, so the setting must live on the
    AgentManifest of each sub-agent role.
  - devteam: expand existing mcp_servers = ["github"] to include
    memory / git / filesystem; populate skills with the expected
    dev-team expertise (replacing the placeholder skills = []).
  - wiki: replace placeholder mcp_servers = [] with [memory, fetch,
    filesystem]. Hand-level skills stays [].
  - lead: hand-level skills was originally [email-writer, writing-
    coach, interview-prep]; interview-prep is for job-interview
    preparation, not lead generation. Replaced with data-analyst
    (used by the qualification-scoring step in the prompt).

schema.toml: register mcp_servers / skills / max_history_messages on
the agent field schema so machine consumers (RegistrySchema in
librefang-types) see the new top-level fields. The
max_history_messages description now points at
librefang_runtime::agent_loop::DEFAULT_MAX_HISTORY_MESSAGES (60
today) by name, so the schema doesn't go stale when the constant
moves again.

agents/README.md: example block + "Adding a New Agent" checklist
mention the allowlists; max_history_messages example is shown
commented out with a prompt-cache caveat.

Open items
----------

`assistant` (the default user-facing agent) keeps `skills = []`
deliberately. It is the generalist entry point — capping its skill
surface at a small allowlist would defeat its "delegate to any
specialist" job. The trade-off is that this single agent still pays
the full skill-definition load on every turn; operators who want a
strict allowlist for `assistant` can override it after install.

Why not adopt PR #89's approach
-------------------------------

#89 covers similar ground but with three issues this PR avoids:

1. mcp_servers = ["_none"] sentinel. #89's body explicitly notes
   it's pending upstream librefang#4808 (mcp_disabled). Shipping a
   magic-string today means coming back later to clean it up. This
   PR uses real allowlists.
2. max_history_messages = 8 / 12 / 15 / 20. Far below today's
   kernel default (60) and #91's direction for long-workflow hands
   (80–120). Every turn that hits the cap invalidates the cached
   prompt prefix; the cost of cache misses exceeds the saving from
   shorter history. This PR uses 60–120.
3. Doubling max_llm_tokens_per_hour (coder 200k→500k, assistant
   300k→500k) widens the per-agent budget — the opposite direction
   from #87's "reduce per-call cost" goal. Left to the operator's
   instance-specific tuning.

Refs librefang/librefang-registry#87, librefang/librefang-registry#89

agents/README.md

@@ -26,6 +26,17 @@ description = "Expert software engineer. Reads, writes, and analyzes code."
 author = "librefang"
 module = "builtin:chat"              # runtime module — builtin:chat for all current agents
 
+# Resource allowlists — keep these small. The kernel injects every entry's
+# tools/skills into the agent's prompt on every LLM call, so an empty list
+# (the default "all available" semantics) on an instance with many MCP
+# servers and skills can burn 30-60k tokens per turn on definitions the
+# agent never uses.
+mcp_servers = ["memory", "git", "github", "filesystem"]   # [] = all available
+skills = ["rust-expert", "python-expert", "git-expert"]   # [] = all available
+# max_history_messages = 60          # OPT-IN override; omit to inherit kernel default
+                                     # (default rises with the kernel; setting too low
+                                     #  thrashes the prompt cache — measure before pinning)
+
 [metadata.routing]
 aliases = ["write code", "fix bug", "implement feature"]    # exact activation phrases
 weak_aliases = ["refactor", "patch", "code change"]         # keyword hints
@@ -171,8 +182,9 @@ Agents can also be used as base templates in a Hand by setting `base = "coder"`
 2. Set `module = "builtin:chat"` unless you have a custom runtime module.
 3. Write a focused `system_prompt` — clear role definition, methodology, and constraints.
 4. Declare only the tools and capabilities the agent actually needs.
-5. Add `[metadata.routing]` aliases so the router can activate the agent by intent.
-6. Run `python scripts/validate.py`.
-7. Submit a PR.
+5. Set `mcp_servers` and `skills` to the minimum the prompt actually references — leaving them empty falls back to "all available", which on a populated instance bloats every LLM call with tool definitions the agent never uses.
+6. Add `[metadata.routing]` aliases so the router can activate the agent by intent.
+7. Run `python scripts/validate.py`.
+8. Submit a PR.
 
 See [CONTRIBUTING.md](../CONTRIBUTING.md) for the full guide.

agents/academic-researcher/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["research", "academic", "papers", "literature-review", "science"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch", "exa-search", "brave-search"]
+skills = ["technical-writer", "writing-coach", "pdf-reader"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = [
   "academic research",

agents/analyst/agent.toml

@@ -4,6 +4,14 @@ description = "Data analyst. Processes data, generates insights, creates reports
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "sqlite-mcp", "postgresql"]
+skills = ["data-analyst", "sql-analyst", "python-expert"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = ["analytics", "metrics analysis", "report analysis"]
 weak_aliases = ["kpi", "insights", "reporting"]

agents/architect/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["architecture", "design", "planning"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch", "github"]
+skills = ["openapi-expert", "technical-writer"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = [
   "system design",

agents/assistant/agent.toml

@@ -12,6 +12,14 @@ tags = [
   "productivity",
 ]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch", "filesystem"]
+skills = []
+
+
+max_history_messages = 120
 [metadata.routing]
 aliases = [
   "general help",

agents/code-reviewer/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["review", "code-quality", "best-practices"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "github", "git"]
+skills = ["code-reviewer", "rust-expert", "python-expert", "typescript-expert"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = [
   "code review",

agents/coder/agent.toml

@@ -5,6 +5,20 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["coding", "implementation", "rust", "python"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "git", "github", "filesystem"]
+skills = [
+  "rust-expert",
+  "python-expert",
+  "typescript-expert",
+  "git-expert",
+  "shell-scripting",
+]
+max_history_messages = 80
+
+
 [metadata.routing]
 aliases = [
   "write code",

agents/customer-support/agent.toml

@@ -12,6 +12,14 @@ tags = [
   "resolution",
 ]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch"]
+skills = ["email-writer", "writing-coach"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "customer support",

agents/data-scientist/agent.toml

@@ -4,6 +4,14 @@ description = "Data scientist. Analyzes datasets, builds models, creates visuali
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "sqlite-mcp", "postgresql"]
+skills = ["data-analyst", "ml-engineer", "python-expert", "sql-analyst"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = ["data science", "build model", "train model", "machine learning"]
 weak_aliases = ["modeling", "statistics"]

agents/debugger/agent.toml

@@ -4,6 +4,14 @@ description = "Expert debugger. Traces bugs, analyzes stack traces, performs roo
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "git", "github", "sentry"]
+skills = ["rust-expert", "python-expert", "shell-scripting"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = [
   "debug this",

agents/devops-lead/agent.toml

@@ -4,6 +4,23 @@ description = "DevOps lead. Manages CI/CD, infrastructure, deployments, monitori
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "git", "github", "filesystem"]
+skills = [
+  "docker",
+  "kubernetes",
+  "terraform",
+  "ansible",
+  "ci-cd",
+  "helm",
+  "prometheus",
+  "sysadmin",
+]
+max_history_messages = 60
+
+
 [metadata.routing]
 aliases = [
   "ci cd",

agents/doc-writer/agent.toml

@@ -4,6 +4,14 @@ description = "Technical writer. Creates documentation, README files, API docs,
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "filesystem", "fetch", "github"]
+skills = ["technical-writer", "writing-coach", "openapi-expert"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "write documentation",

agents/email-assistant/agent.toml

@@ -12,6 +12,14 @@ tags = [
   "productivity",
 ]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "gmail"]
+skills = ["email-writer", "writing-coach"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "draft email",

agents/health-tracker/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["health", "wellness", "fitness", "medication", "habits", "tracking"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory"]
+skills = []
+skills_disabled = true
+
+
 [metadata.routing]
 aliases = [
   "health tracking",

agents/hello-world/agent.toml

@@ -4,6 +4,14 @@ description = "A friendly greeting agent that can read files, search the web, an
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch"]
+skills = []
+skills_disabled = true
+
+
 [metadata.routing]
 aliases = [
   "hello world",

agents/home-automation/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["smart-home", "iot", "automation", "devices", "monitoring", "home"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch"]
+skills = []
+skills_disabled = true
+
+
 [metadata.routing]
 aliases = [
   "smart home",

agents/legal-assistant/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["legal", "contracts", "compliance", "research", "review", "documents"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch"]
+skills = ["compliance", "writing-coach"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = [
   "legal review",

agents/meeting-assistant/agent.toml

@@ -12,6 +12,14 @@ tags = [
   "productivity",
 ]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "google-calendar", "gmail"]
+skills = ["writing-coach"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "meeting notes",

agents/ops/agent.toml

@@ -4,6 +4,20 @@ description = "DevOps agent. Monitors systems, runs diagnostics, manages deploym
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "git"]
+skills = [
+  "docker",
+  "sysadmin",
+  "shell-scripting",
+  "linux-networking",
+  "prometheus",
+]
+max_history_messages = 60
+
+
 [metadata.routing]
 aliases = [
   "service status",

agents/orchestrator/agent.toml

@@ -4,6 +4,14 @@ description = "Meta-agent that decomposes complex tasks, delegates to specialist
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory"]
+skills = []
+
+
+max_history_messages = 120
 [metadata.routing]
 aliases = [
   "multi agent",

agents/personal-finance/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["finance", "budget", "expenses", "savings", "planning", "money"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory"]
+skills = ["data-analyst"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "budget planning",

agents/planner/agent.toml

@@ -4,6 +4,14 @@ description = "Project planner. Creates project plans, breaks down epics, estima
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "github", "linear", "jira"]
+skills = ["project-manager"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "project plan",

agents/recipe-assistant/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["cooking", "recipes", "meal-planning", "nutrition", "food"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch"]
+skills = []
+skills_disabled = true
+
+
 [metadata.routing]
 aliases = [
   "recipe helper",

agents/recruiter/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["recruiting", "hiring", "resume", "outreach", "talent", "hr"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "gmail", "linear"]
+skills = ["email-writer", "writing-coach", "interview-prep"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "candidate screening",

agents/researcher/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["research", "analysis", "web"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch", "exa-search", "brave-search"]
+skills = ["technical-writer", "writing-coach"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = [
   "web research",

agents/sales-assistant/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["sales", "crm", "outreach", "pipeline", "prospecting", "deals"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "gmail", "slack"]
+skills = ["email-writer", "writing-coach"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = ["sales outreach", "crm update", "pipeline review", "deal tracking"]
 weak_aliases = ["crm", "leads"]

agents/security-auditor/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["security", "audit", "vulnerability"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "github", "git"]
+skills = ["security-audit", "compliance", "oauth-expert"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = [
   "security audit",

agents/social-media/agent.toml

@@ -12,6 +12,14 @@ tags = [
   "analytics",
 ]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch"]
+skills = ["writing-coach"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "social media plan",

agents/test-engineer/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["testing", "qa", "validation"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "github", "git"]
+skills = ["api-tester", "python-expert", "rust-expert", "typescript-expert"]
+
+
+max_history_messages = 80
 [metadata.routing]
 aliases = [
   "test plan",

agents/translator/agent.toml

@@ -12,6 +12,14 @@ tags = [
   "i18n",
 ]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch"]
+skills = ["writing-coach"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "translate document",

agents/travel-planner/agent.toml

@@ -5,6 +5,14 @@ author = "librefang"
 module = "builtin:chat"
 tags = ["travel", "planning", "itinerary", "booking", "logistics", "vacation"]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch", "google-maps", "google-calendar"]
+skills = []
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "trip itinerary",

agents/tutor/agent.toml

@@ -12,6 +12,14 @@ tags = [
   "knowledge",
 ]
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch"]
+skills = ["writing-coach", "python-expert"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "teach me",

agents/writer/agent.toml

@@ -4,6 +4,14 @@ description = "Content writer. Creates documentation, articles, and technical wr
 author = "librefang"
 module = "builtin:chat"
 
+# Per-agent resource allowlists (refs librefang/librefang-registry#87).
+# Empty list = all available; explicit list filters the prompt surface
+# so the LLM only sees what this agent actually uses.
+mcp_servers = ["memory", "fetch"]
+skills = ["technical-writer", "writing-coach"]
+
+
+max_history_messages = 60
 [metadata.routing]
 aliases = [
   "write article",

hands/analytics/HAND.toml

@@ -23,6 +23,12 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "sqlite-mcp", "postgresql", "filesystem"]
+skills = ["data-analyst", "sql-analyst", "python-expert", "data-pipeline"]
+
+
 [routing]
 aliases = [
   "data analysis",

hands/apitester/HAND.toml

@@ -24,6 +24,18 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "github", "fetch"]
+skills = [
+  "api-tester",
+  "openapi-expert",
+  "graphql-expert",
+  "python-expert",
+  "typescript-expert",
+]
+
+
 [[requires]]
 key = "curl"
 label = "curl must be installed"

hands/browser/HAND.toml

@@ -28,6 +28,12 @@ tools = [
   "file_read",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "puppeteer"]
+skills = []
+
+
 [routing]
 aliases = [
   "open website",

hands/clip/HAND.toml

@@ -17,6 +17,15 @@ tools = [
   "memory_recall",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "filesystem"]
+skills = []
+# `skills_disabled = true` lives on each [agents.*] below — HandDefinitionRaw
+# in librefang-hands doesn't carry the field, so a hand-level entry would be
+# silently dropped by serde.
+
+
 [routing]
 aliases = [
   "clip video",
@@ -219,6 +228,7 @@ description = "AI video editor — downloads, transcribes, and creates viral sho
 module = "builtin:chat"
 provider = "default"
 model = "default"
+skills_disabled = true
 max_tokens = 8192
 temperature = 0.4
 max_iterations = 40
@@ -622,6 +632,7 @@ description = "Content writer. Creates scripts, captions, titles, and descriptio
 module = "builtin:chat"
 provider = "default"
 model = "default"
+skills_disabled = true
 max_tokens = 4096
 temperature = 0.7
 system_prompt = """You are Writer, a short-form video content specialist within the Clip Hand.
@@ -760,6 +771,7 @@ description = "Social media strategist. Plans distribution, scheduling, and enga
 module = "builtin:chat"
 provider = "default"
 model = "default"
+skills_disabled = true
 max_tokens = 4096
 temperature = 0.7
 system_prompt = """You are Distributor, the publishing and distribution specialist within the Clip Hand.

hands/collector/HAND.toml

@@ -25,6 +25,12 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "fetch", "exa-search", "brave-search", "filesystem"]
+skills = ["data-analyst"]
+
+
 [routing]
 aliases = [
   "monitor changes",

hands/creator/HAND.toml

@@ -20,6 +20,15 @@ tools = [
   "memory_recall",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "filesystem"]
+skills = []
+# `skills_disabled = true` lives on each [agents.*] below — HandDefinitionRaw
+# in librefang-hands doesn't carry the field, so a hand-level entry would be
+# silently dropped by serde.
+
+
 [routing]
 aliases = [
   "generate image",
@@ -172,6 +181,7 @@ description = "AI media studio — generates images, videos, music, and speech f
 module = "builtin:chat"
 provider = "default"
 model = "default"
+skills_disabled = true
 max_tokens = 8192
 temperature = 0.5
 max_iterations = 30
@@ -279,6 +289,7 @@ description = "Prompt engineer that crafts detailed, effective prompts for media
 module = "builtin:chat"
 provider = "default"
 model = "default"
+skills_disabled = true
 max_tokens = 4096
 temperature = 0.8
 system_prompt = """You are Prompt Writer, a creative prompt engineer within the Creator Hand.

hands/devops/HAND.toml

@@ -24,6 +24,30 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = [
+  "memory",
+  "git",
+  "github",
+  "filesystem",
+  "sentry",
+  "elasticsearch",
+]
+skills = [
+  "docker",
+  "kubernetes",
+  "terraform",
+  "ansible",
+  "ci-cd",
+  "helm",
+  "prometheus",
+  "sysadmin",
+  "linux-networking",
+  "shell-scripting",
+]
+
+
 [[requires]]
 key = "curl"
 label = "curl must be installed"

hands/devteam/HAND.toml

@@ -30,11 +30,20 @@ tools = [
   "workflow_run",
 ]
 
-# MCP servers: all agents can access these (per-agent mcp_servers further restricts)
-mcp_servers = ["github"]
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "github", "git", "filesystem"]
+skills = [
+  "github",
+  "git-expert",
+  "rust-expert",
+  "python-expert",
+  "typescript-expert",
+  "code-reviewer",
+  "api-tester",
+  "ci-cd",
+]
 
-# Skills: all available (agents can restrict individually)
-skills = []
 
 # Plugins: useful for dev workflow
 allowed_plugins = ["todo-tracker", "auto-summarizer", "episodic-memory"]

hands/lead/HAND.toml

@@ -24,6 +24,19 @@ tools = [
   "knowledge_query",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = [
+  "memory",
+  "fetch",
+  "gmail",
+  "linear",
+  "exa-search",
+  "brave-search",
+]
+skills = ["email-writer", "writing-coach", "data-analyst"]
+
+
 [routing]
 aliases = [
   "lead generation",

hands/linkedin/HAND.toml

@@ -23,6 +23,12 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "fetch"]
+skills = ["writing-coach", "email-writer"]
+
+
 [routing]
 aliases = ["linkedin", "profile optimization", "professional networking"]
 weak_aliases = ["professional engagement", "linkedin post"]

hands/predictor/HAND.toml

@@ -23,6 +23,12 @@ tools = [
   "knowledge_query",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "fetch", "exa-search", "brave-search"]
+skills = ["data-analyst", "python-expert"]
+
+
 [routing]
 aliases = [
   "predict",

hands/reddit/HAND.toml

@@ -23,6 +23,12 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "fetch"]
+skills = ["writing-coach"]
+
+
 [routing]
 aliases = ["reddit", "subreddit", "reddit post", "reddit monitor"]
 weak_aliases = ["karma", "reddit thread"]

hands/researcher/HAND.toml

@@ -26,6 +26,12 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "fetch", "exa-search", "brave-search"]
+skills = ["technical-writer", "writing-coach", "python-expert", "pdf-reader"]
+
+
 [routing]
 aliases = [
   "deep research",

hands/strategist/HAND.toml

@@ -24,6 +24,12 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "fetch", "exa-search", "brave-search"]
+skills = ["project-manager", "data-analyst", "writing-coach"]
+
+
 [routing]
 aliases = [
   "strategic analysis",

hands/trader/HAND.toml

@@ -25,6 +25,12 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "fetch", "postgresql", "sqlite-mcp"]
+skills = ["data-analyst", "sql-analyst", "python-expert"]
+
+
 [routing]
 aliases = [
   "trade",

hands/twitter/HAND.toml

@@ -24,6 +24,12 @@ tools = [
   "event_publish",
 ]
 
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "fetch"]
+skills = ["writing-coach"]
+
+
 [routing]
 aliases = [
   "twitter",

hands/wiki/HAND.toml

@@ -18,8 +18,12 @@ tools = [
   "memory_store",
 ]
 
-mcp_servers = []
+# Per-hand resource allowlists (refs librefang/librefang-registry#87).
+# Inherited by every [agents.*] in this hand unless overridden.
+mcp_servers = ["memory", "fetch", "filesystem"]
 skills = []
+
+
 allowed_plugins = []
 # memory_store is used ONLY for dashboard metrics.
 # No wiki content enters LibreFang's general memory — all knowledge lives in the vault.

schema.toml

@@ -204,6 +204,26 @@ required = true
 description = "Runtime module (builtin:chat, builtin:tool, etc.)"
 example = "builtin:chat"
 
+[agent.fields.mcp_servers]
+type = "array"
+required = false
+description = "MCP server allowlist. Empty list (default) means every connected MCP server's tools are injected into this agent's prompt — on a typical instance that is ~85 tool definitions and ~50k tokens per call. Set an explicit list to trim the surface to only the servers this agent actually uses (e.g. [\"memory\", \"git\", \"github\"])."
+item_type = "string"
+example = ["memory", "git", "github"]
+
+[agent.fields.skills]
+type = "array"
+required = false
+description = "Skill allowlist. Empty list (default) means every installed skill is available. Set an explicit list to load only what this agent needs."
+item_type = "string"
+example = ["rust-expert", "python-expert"]
+
+[agent.fields.max_history_messages]
+type = "number"
+required = false
+description = "Per-agent override for the message-history trim cap. Omit (default) inherits from KernelConfig.max_history_messages, which itself falls back to the compiled constant librefang_runtime::agent_loop::DEFAULT_MAX_HISTORY_MESSAGES (60 today). Values below 4 are clamped at runtime with a warning. Pinning this *below* the kernel default trashes the prompt cache — every turn that hits the cap invalidates the cached prefix and costs more than the smaller history saves. Set this only to raise the cap for iteration-heavy agents that legitimately need more context, or to lower it for agents whose conversations are demonstrably short and where you've measured the trade-off."
+example = 80
+
 # --- Agent: [model] section ---
 
 [agent.sections.model]