ccs-align

Run the CCS Align seat's hourly breathing cycle — prove the local claude-mem worker is healthy, pull needle observations through search → timeline → get_observations, land them in a seat-owned middle cache via atomic grab → append → filter exclude-marks → replace, manage exclude marks, and walk house → project → seat rules to detect conflicts (SHADOW_HOUSE, DENY_ALLOW, DRIFT, CLOCK_HEADER) with an append-only rules-report.md. Use when asked to run CCS Align, breathe the alignment seat, refresh the middle cache, exclude or restore an observation, walk rules, check rules conflicts, or check the Worker Watch board.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%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.

The body is highly actionable with executable commands, locked data shapes, and a well-sequenced runbook with validation checkpoints. It is held back by repetition of the forbids/boundary material across multiple sections, time-sensitive version chatter, and a monolithic structure that does not split detail into reference files.

Suggestions

Consolidate the repeated boundary/forbids content into a single 'What this is NOT / Hard forbids' section and remove the duplicates from the intro, 'Later phases', and elsewhere.

Move the verbose Phase 1 exclude-marks JSON schema and Phase 2 layer-walk tables into reference files under references/, leaving SKILL.md as a lean overview with one-level-deep pointers.

Push the worker-plugin version-lag (13.24.5 vs 13.24.1) detail into an ops/MISS section or the plan file rather than the main body, since it is time-sensitive and will go stale.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and free of concept-explanation padding, but the boundary/forbids content ('not a context compiler, not a brainbeat', the DELETE/rewrite forbids) is repeated across 'What this is NOT', 'Hard forbids', and 'Later phases', and time-sensitive version/date detail (13.24.5 vs 13.24.1, 'As of this sign-off') adds noise that could be trimmed.	3 / 5
Actionability	Fully executable guidance throughout — real curl commands using $WORKER_PORT, a copy-paste port-resolution snippet, locked JSON record shapes, exact import paths, and concrete TS invocations of the helper functions covering the common cases.	5 / 5
Workflow Clarity	The hourly cycle is a numbered 10-step runbook expanded per-step with explicit validation checkpoints (health → abort with a one-line miss; 'If health cannot be proven, record a MISS and stop'), dedupe/exclude verification feedback loops, and a Phase 3 sign-off checklist — matching the clear-sequence-with-validation anchor.	5 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets absent), and the body is a single ~438-line monolithic document that inlines Phase 1/Phase 2 operational detail and full JSON shapes which could live in one-level reference files; the only external pointers are repo source paths and a plan link rather than clearly signaled skill references.	3 / 5
	Total	16 / 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, specific description that concretely enumerates the seat's actions and provides an explicit, multi-phrase 'Use when' trigger clause. Its only weak spot is reliance on project-internal jargon for some trigger terms.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions with named specifics — 'pull needle observations through search → timeline → get_observations', 'atomic grab → append → filter exclude-marks → replace', and the four named conflict classes (SHADOW_HOUSE, DENY_ALLOW, DRIFT, CLOCK_HEADER) — giving comprehensive coverage.	5 / 5
Completeness	Clearly answers both 'what' (the hourly breathing cycle, pulling observations, atomic middle-cache landing, exclude marks, rules walking) and 'when' via an explicit 'Use when...' clause with concrete trigger phrases, matching the 5 anchor.	5 / 5
Trigger Term Quality	The 'Use when asked to run CCS Align, breathe the alignment seat, refresh the middle cache, exclude or restore an observation, walk rules, check rules conflicts, or check the Worker Watch board' clause gives good natural-term coverage with some synonyms, though several terms are project-internal jargon a general user would not spontaneously say.	4 / 5
Distinctiveness Conflict Risk	Tightly bound to a specific niche — 'CCS Align seat', 'local claude-mem worker', 'Worker Watch board' — with distinct triggers and minimal overlap risk against other skills.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: thedotmack/claude-mem
Path: plugin/skills/ccs-align/SKILL.md
Commit: d8bc975

Reviewed: 4 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.