skill-composer

A skill-set manager for AI coding agents. It loads only the skills relevant to the current task and checks that they are approved and authorized.

In plain words
What is it for?
It helps configure an agent for a specific project, enforce approved skill use, and manage limits such as context size and memory use.
Why use it?
It reduces unnecessary context and permissions while keeping skill choices traceable to project requirements. It also applies governance checks before skills are composed.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/layer1labs/specsmith/skill-composer
Any agent
npx skills add layer1labs/specsmith --skill skill-composer
Clone the repo
git clone --depth 1 https://github.com/layer1labs/specsmith

Made for: Claude Code, Codex.

Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 519 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00519
Opus 5 $0.00000 $0.00260
Sonnet 5 $0.00000 $0.00104
Haiku 4.5 $0.00000 $0.00052

Measured 2d ago against content hash f934096a1035, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

skill-composer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/skills/skill-composer/SKILL.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

SKILL.md

Skill Composer

Requirements Covered

Governance Requirements

  • REQ-012: Least-Privilege Agent Permissions
  • REQ-217: Agent Permissions Check
  • REQ-220: Policy Guardrails
  • REQ-244: Context Window Sizing
  • REQ-245: Live Context Fill Indicator
  • REQ-246: Auto Context Compression
  • REQ-247: Hard Context Ceiling

Core Principle

The skill-composer skill enables agents to dynamically compose and manage their skill sets based on the current task, project context, and governance requirements. It ensures that agents only load the necessary skills for their current operation while maintaining compliance with Specsmith's governance framework.

Skill Composition Rules

1. Context-Aware Skill Loading

  • Load only skills relevant to the current task
  • Dynamically adjust skill set based on project requirements
  • Ensure all loaded skills are properly validated and authorized

2. Governance Compliance

  • All skill compositions must pass preflight checks
  • Skills must be from the approved skill catalog
  • Any skill composition must be traceable to a work item or requirement

3. Performance Optimization

  • Minimize memory footprint by loading only necessary skills
  • Optimize skill loading order for maximum efficiency
  • Cache skill compositions when appropriate

Implementation Details

Skill Selection Algorithm

  1. Analyze the current task requirements
  2. Identify required capabilities from the project's governance rules
  3. Select appropriate skills from the available catalog
  4. Validate that all selected skills are compliant with current governance
  5. Compose the final skill set for execution

Skill Validation

  • Verify that each skill meets its documented requirements
  • Ensure skills are compatible with the current project context
  • Confirm that skill combinations don't violate governance constraints

Configuration

Default Behavior

By default, the skill-composer will:

  • Load core governance skills (preflight-gate, governed-agent-loop)
  • Select task-specific skills based on the current operation
  • Apply context-aware optimization

Read the full file on GitHub · 71 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 71 lines · 0 tokens per session scan A f934096a1035

Subscribe to this mod's changes

skill-composer is a skill published in the GitHub repository layer1labs/specsmith (7 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 519 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

arch-audit

Audit Claude Code architecture files against Anthropic docs and release notes, and verify internal ecosystem consistency. Run on demand to maintain compliance, catch new features, and keep the context system clean.

marcoguillermaz/Tierward · 41 tokens

skill-dev

Code quality audit: detect cross-module coupling, N+1 queries, dead exports, antipatterns, over-large components. Cross-checks against refactoring backlog.

marcoguillermaz/Tierward · 36 tokens

accessibility-audit

Unified accessibility audit - axe-core WCAG 2.2 scan, APCA contrast measurement, static a11y checks (aria-label, tabindex, form labels, focus visibility, onClick on non-interactive). Static and live modes.

marcoguillermaz/Tierward · 53 tokens

pr-review

Autonomous local code review. Reviews either an open PR's diff (via gh) or the local working diff against a base branch (--local, auto-selected when no PR exists yet), spawns a dedicated review subagent with universal + stack-specific severity criteria, posts the review as a comment on the PR (audit trail; skipped in…

marcoguillermaz/Tierward · 167 tokens

responsive-audit

Responsive audit: test pages at 375/768/1024px breakpoints via Playwright. Checks overflow, tap targets, sidebar collapse, text reflow, WCAG 1.4.4 zoom.

marcoguillermaz/Tierward · 47 tokens

security-audit

Security audit: auth/authz on API routes, input validation, row-level access control (RLS in Postgres, ORM scopes or app-level guards elsewhere), response shape review, secret exposure, HTTP headers. Native mode checks entitlements and Keychain. MCP-aware (v1.20+): when mcp-nvd server is wired, Step 3c queries live…

marcoguillermaz/Tierward · 109 tokens