agent-hooks

Four lifecycle hooks every agent runs: on-session-start scans LESSONS-LEARNED.md, resumes checkpoints; on-pre-delegate verifies tracker issues, file partitions; on-post-delegate runs fast-review, CI checks; on-session-end runs the health checks, writes logs. Use when starting a new session, running pre-flight checks before delegation, coordinating between agents, reviewing a completed handoff, or wrapping up a session. Trigger terms: multi-agent setup, delegate tasks, agent coordination, session management, run pre-flight checks, start a new session, coordinate between agents, wrap up session

Quality

90%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

85%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A concise, highly actionable body with clear sequenced workflows and real validation gates on the destructive/batch hooks. The main weakness is progressive disclosure: the body points to HOOKS-REFERENCE.md three times, but that file (and any bundle directory) is absent.

Suggestions

Ship the referenced HOOKS-REFERENCE.md (or create a references/ bundle) so the three 'See HOOKS-REFERENCE.md' links resolve — dead reference links break navigation and undercut the skill's progressive-disclosure design.

Replace the `opencastle log --type session ...` placeholder with a concrete example invocation so on-session-end step 2 is copy-paste ready.

Add a one-line note clarifying that the hook tables are the complete checklist vs. HOOKS-REFERENCE.md's extended examples, so readers know what is inline vs. external without guessing.

Dimension	Reasoning	Score
Conciseness	The body is lean tables of executable commands with no padding or explanations of concepts Claude already knows; the one-line intro ('Conventions (not auto-triggers)...') earns its place as necessary context, matching the lean-and-efficient anchor.	5 / 5
Actionability	Concrete, executable commands dominate (`rg -n`, `cat`, `opencastle doctor`, `gh issue view`, `comm -12`, `pnpm lint && pnpm typecheck && pnpm test`), but template placeholders (`opencastle log --type session ...`, `TAS-XX`) leave minor gaps versus fully copy-paste-ready guidance.	4 / 5
Workflow Clarity	Each hook is a numbered sequence with explicit validation gates ('All 6 must pass before runSubagent', the on-session-end HARD GATE) and feedback loops (3rd failure -> AGENT-FAILURES.md, re-delegate), satisfying the explicit-checkpoints anchor for the destructive/batch hooks.	5 / 5
Progressive Disclosure	References to HOOKS-REFERENCE.md are well-signaled and one level deep, but no bundle directory or HOOKS-REFERENCE.md file exists, so the navigation links are dead ends — the structure is present but the referenced detail content is missing.	3 / 5
	Total	17 / 20 Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, well-structured description that covers what each hook does, when to use them, and lists natural trigger terms with synonyms. Only minor overlap risk on a few broad coordination terms keeps distinctiveness from the top anchor.

Dimension	Reasoning	Score
Specificity	Lists eight-plus concrete actions across the four hooks ('scans LESSONS-LEARNED.md, resumes checkpoints', 'verifies tracker issues, file partitions', 'runs fast-review, CI checks', 'runs the health checks, writes logs'), matching the comprehensive-coverage anchor.	5 / 5
Completeness	It explicitly answers both what (the four hooks and their actions) and when ('Use when starting a new session, running pre-flight checks before delegation...'), with concrete trigger phrases — the top anchor.	5 / 5
Trigger Term Quality	The explicit 'Trigger terms:' clause provides natural phrases with synonyms ('start a new session'/'session management', 'delegate tasks'/'coordinate between agents', 'wrap up session'), matching the comprehensive-synonym anchor.	5 / 5
Distinctiveness Conflict Risk	The agent-lifecycle-hooks niche is distinct with specific triggers, but broad terms like 'session management' and 'agent coordination' create minor overlap risk with general multi-agent coordination skills, so it sits just below the minimal-conflict anchor.	4 / 5
	Total	19 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 3 missing	Warning

	Total	15 / 16 Passed

Repository: monkilabs/opencastle
Path: src/orchestrator/skills/agent-hooks/SKILL.md
Commit: fa0c341

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.