CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

memory-ci-capture

Internal CI skill: synthesize GitHub delivery context into a Basic Memory project update. Use after `bm ci collect`; return only AgentSynthesis JSON for `bm ci publish`.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

92%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 a model instruction-only skill: exact input location, a complete output contract, per-field synthesis rules, and event-specific guidance for the two common cases (merged PRs, production deploys), all with strong anti-hallucination guardrails. The only trimmable fat is the brief motivational tagline, which keeps conciseness just short of perfect.

DimensionReasoningScore

Conciseness

The body is lean with no explanations of concepts Claude already knows, but the three-sentence tagline ("GitHub records the mechanics. Basic Memory remembers what changed and why.") and the "not marketing copy" aside are minor flavor a machine-invoked CI skill could trim. This fits anchor 4 (efficient with minor trimmable instances) better than anchor 5's every-token-earns-its-place.

4 / 5

Actionability

Guidance is fully executable for an instruction-only skill: the exact input path `.github/basic-memory/project-update-context.json`, a complete copy-paste-ready 8-field `AgentSynthesis` JSON template, per-field rules, and explicit edge handling ("If evidence is absent, say so briefly in `verification` or leave the field empty"). Per the scoring notes, the absence of code is not penalized when guidance is this concrete.

5 / 5

Workflow Clarity

The single action is unambiguous: read the context file, apply the synthesis rules, return the JSON, with anti-hallucination checkpoints ("Do not invent tests, deploy checks, linked issues, product impact, or decisions"). It is not a destructive or batch operation, so the validation cap does not apply, and the simple-skill exception for a clear single-task skill holds.

5 / 5

Progressive Disclosure

This is a short (~50-line) single-purpose skill with well-organized sections (Inputs, Output, Synthesis Rules, Event Guidance); no bundle files exist and none are needed. Under the simple-skill guideline, well-organized sections with no external references warranted is a top score.

5 / 5

Total

19

/

20

Passed

Description

66%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 tight, honest description with an explicit what-and-when structure and virtually no conflict risk thanks to its tool-specific vocabulary. Its main weakness is trigger and capability coverage: it relies on a single CLI-command trigger and jargon-heavy terms rather than natural phrases, and it does not enumerate the synthesis capabilities the skill provides.

Suggestions

Enumerate the concrete outputs the synthesis produces (e.g., change summaries, user-facing vs. internal changes, decision and task candidates) so the 'what' is comprehensive rather than a single action.

Broaden trigger coverage beyond the single `bm ci collect` condition with natural phrases such as 'Use when capturing a merged PR or production deploy as project memory', so the skill triggers on delivery-memory requests phrased in ordinary language.

Add one or two natural synonyms alongside the CLI jargon (e.g., 'GitHub delivery context (merged PRs, production deploys)') to improve natural keyword coverage without inflating length.

DimensionReasoningScore

Specificity

"synthesize GitHub delivery context into a Basic Memory project update" and "return only AgentSynthesis JSON" name the domain plus 1-2 concrete actions, but the description never enumerates what the synthesized update contains (summaries, decisions, tasks), so coverage is not comprehensive. It falls short of anchor 4, which expects several listed specific actions with only minor gaps.

3 / 5

Completeness

The description explicitly answers both parts: what ("synthesize GitHub delivery context into a Basic Memory project update") and when ("Use after `bm ci collect`"). It is not a 3 because the "when" is an explicit trigger rather than weakly implied, and not a 5 because it offers only that single trigger condition without concrete trigger phrases covering variations.

4 / 5

Trigger Term Quality

Keywords like "bm ci collect", "bm ci publish", "AgentSynthesis JSON", and "GitHub delivery context" are relevant and specific, but the mix is dominated by CLI/toolchain jargon with no natural-language variations or synonyms. This matches anchor 3 (some relevant keywords, missing common variations) rather than anchor 4's good natural-term coverage.

3 / 5

Distinctiveness Conflict Risk

"Internal CI skill", "Basic Memory", and the `bm ci` command names carve out a clear niche with distinct, toolchain-specific triggers, so the risk of firing for the wrong skill is minimal. This is a direct match for anchor 5.

5 / 5

Total

15

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

basicmachines-co/basic-memory

Path

skills/memory-ci-capture/SKILL.md

Reviewed

Table of Contents

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.