Name: learn
Rating: 72 (1 reviews)
Author: tdg-ninja

learn

Update the project's long-term memory after a merge to main. Reads the merged diff and the current memory, then writes Expert shards, discovered invariants, candidate lints, and AGENTS.md pointers — all on a reviewable learn/<sha> PR. Use post-merge (the harness invokes it automatically) or with --rebuild to regenerate memory from scratch. Triggers - learn, expert-update, update memory, update expert, post-merge memory, self-improve (project)

Quality

88%

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

The canonical home for this skill is learn in tdg-ninja/context-specs-factory-ai

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 well-structured, opinionated skill body that sequences a multi-step post-merge memory workflow with explicit validation and feedback loops, and cleanly offloads detail to one-level-deep reference files. It is mostly lean, with some repeated framing and a few abstract steps keeping conciseness and actionability just short of perfect.

Suggestions

De-duplicate the 'lint is the highest-value memory / a rule the agent cannot ship past' framing — it appears in the intro, P6, and Hard nevers; state it once and reference it.

Make Step 4's shard and AGENTS.md updates more concrete (e.g., a one-line example of a pointer edit or shard diff) so 'apply diffs to the relevant shard(s)' is executable rather than abstract.

Tighten the philosophy section (P1–P8) by pruning editorializing adjectives; the principles are valuable but several could be stated in fewer tokens.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's competence (no basic explanations of git/PRs/memory), but repeats the lint-value framing across the intro, P6, and 'Hard nevers' and includes some editorializing that could be trimmed.	4 / 5
Actionability	Provides concrete scripts and commands (scripts/check-agents-md.sh, scripts/local-checks.sh, claude -p "/learn --since <sha> --sha <sha>", git ls-remote) with a sequenced Step 0–6 flow, but a few steps like 'apply diffs to the relevant shard(s)' stay abstract rather than copy-paste ready.	4 / 5
Workflow Clarity	Steps 0–6 are clearly sequenced with explicit validation in Step 5 (run check-agents-md.sh, re-run drafted lint) and a real feedback loop — 'The drafted lint MUST pass against the just-merged code before you include it — run it; if it fails on current main it's wrong' — plus idempotency checks in Step 0.	5 / 5
Progressive Disclosure	The SKILL.md is an overview that points to five one-level-deep references (routing-rules.md, expert-shards.md, agents-md-guidance.md, invariant-discovery.md, consensus.md), each clearly signaled with 'See references/X.md'; detailed rules are appropriately split into those files, all verified to exist.	5 / 5
	Total	18 / 20 Passed

Description

92%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, third-person description that concretely names the skill's actions and output surfaces, states when to use it, and lists natural trigger terms. Minor room to broaden trigger synonyms, but it clearly distinguishes itself and answers both what and when.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Reads the merged diff and the current memory, then writes Expert shards, discovered invariants, candidate lints, and AGENTS.md pointers' and 'regenerate memory from scratch' — giving comprehensive coverage of what the skill does.	5 / 5
Completeness	Explicitly answers both what (the read/write actions) and when ('Use post-merge (the harness invokes it automatically) or with --rebuild to regenerate memory from scratch') with concrete trigger phrases.	5 / 5
Trigger Term Quality	The 'Triggers - learn, expert-update, update memory, update expert, post-merge memory, self-improve (project)' list gives good natural keyword coverage, but lacks synonyms and file-extension variants that would push it to a 5.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (post-merge long-term memory reconciliation via learn/<sha> PRs) with distinct triggers, making overlap with other skills minimal.	5 / 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
referenced_paths_exist	Referenced path issues: 5 missing	Warning

	Total	15 / 16 Passed

Repository: tdg-ninja/context-specs-claude-code
Path: skills/harness/learn/SKILL.md
Commit: 3a7a725

Reviewed: about 19 hours 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.