CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

conversation-hygiene

Triage and archive accumulated Codex conversations without losing wanted work. Use when cleaning up old Codex threads, deciding which conversations still need attention, or resuming a large conversation cleanup. Do not use for ordinary backlog capture, feature implementation, or issue-tracker cleanup.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

81%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 content is a well-structured, highly actionable instruction-only skill with an unusually strong validation posture for a batch operation — verification, retry, and failure-handling steps appear at every phase. Its weaknesses are mild: some repetition of checkpoint guidance across sections, an unnamed tool interface, and no use of reference files for a 100+ line body.

Suggestions

Name the specific Codex thread tools (or at least one concrete invocation example) in 'Gather evidence in bounded batches' so the operating interface is unambiguous rather than the generic "Codex thread tools".

Consolidate the checkpoint-field and resume guidance into one place (it currently recurs in 'Gather evidence in bounded batches', 'Account for each outcome', and 'Execute and close') to cut repetition and tighten conciseness.

Consider moving the backlog item template and its lifecycle rules into a short reference file (e.g., references/backlog-item.md) to keep SKILL.md a lean overview and strengthen progressive disclosure for the 100+ line body.

DimensionReasoningScore

Conciseness

The body is dense with project-specific policy and never explains concepts Claude already knows, but checkpoint guidance recurs across three sections ("Gather evidence", "Account for each outcome", "Execute and close") and some sentences are abstract principle statements ("A task and an outcome have separate lifecycles...") that could be tightened. Efficient with minor trim opportunities — anchor 4, not 5.

4 / 5

Actionability

Concrete, actionable throughout: a copy-paste backlog item template, an explicit checkpoint field list ("scope, inventory progress, reviewed IDs and reasons, preservation destinations..."), per-case decision branches, and a required reporting checklist ("reviewed, archived, kept active, unresolved, failed, and not-yet-reviewed counts"). Minor gap: the operating interface is only ever "Codex thread tools" with no named tools or commands, keeping it just below fully executable.

4 / 5

Workflow Clarity

Four clearly sequenced phases (gather evidence → account for outcomes → propose → execute) with explicit validation and error-recovery loops for a batch operation: "recheck each authorized task for new activity", "confirm the write succeeded", "verify each in archived-task state", "If preservation fails, leave its source task unarchived", "Do not infer success from a mutation response alone". The batch-operation cap at 3 does not apply because validation is pervasive; this matches the anchor-5 pattern of sequence plus feedback loops plus checklists.

5 / 5

Progressive Disclosure

Well-organized single file with clear section headers and no nested or buried references; nothing that clearly belongs in a separate file is inlined. However, at ~118 lines the under-50-line no-references exception does not apply, and no one-level-deep reference file splits out the backlog template or lifecycle rules, so it sits at anchor 4 rather than 5.

4 / 5

Total

17

/

20

Passed

Description

87%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.

The description is strong: it states concrete actions, gives explicit multi-scenario 'Use when' triggers plus negative triggers, and carves out a distinct niche with low conflict risk. The only improvements are adding natural synonyms like "tasks" or "sessions" to the trigger terms and slightly broadening the action coverage.

DimensionReasoningScore

Specificity

Names the domain and several concrete actions — "Triage and archive accumulated Codex conversations without losing wanted work", "deciding which conversations still need attention", "resuming a large conversation cleanup" — matching the anchor for several specific actions with minor gaps. Not 5 because it omits parts of the workflow (e.g., preservation destinations, verification), and not 3 because it goes well beyond 1-2 actions.

4 / 5

Completeness

Explicitly answers both what ("Triage and archive accumulated Codex conversations without losing wanted work") and when ("Use when cleaning up old Codex threads, deciding which conversations still need attention, or resuming a large conversation cleanup"), with concrete trigger phrases and negative triggers ("Do not use for..."). This matches the anchor-5 example structure; score 4 would require the 'when' to be less explicit, which it is not.

5 / 5

Trigger Term Quality

Includes natural phrases users would say: "cleaning up old Codex threads", "conversations", "conversation cleanup", "archive". Not 5 because common synonyms such as "tasks", "sessions", or "clear out" are missing, which the body's own vocabulary ("tasks") shows users would likely use.

4 / 5

Distinctiveness Conflict Risk

Clear niche (Codex conversation/task cleanup) with explicit exclusions — "Do not use for ordinary backlog capture, feature implementation, or issue-tracker cleanup" — that sharply reduce overlap with adjacent skills. Minimal conflict risk, matching anchor 5; anchor 4 would imply residual overlap with closely related skills, which the negative triggers largely rule out.

5 / 5

Total

18

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

relative_links

Relative link issues: 1 suspicious

Warning

Total

15

/

16

Passed

Repository

EpicenterHQ/epicenter

Path

.agents/skills/conversation-hygiene/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.