CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

speckit-converge

Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.

Quality

70%

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

Fix and improve this skill with Tessl

tessl review fix ./apps/reference/.claude/skills/speckit-converge/SKILL.md

The canonical home for this skill is speckit-converge in RihardsJ/software-factory

SKILL.md
Quality
Evals
Security

Quality

Content

70%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 an exceptionally rigorous, actionable workflow with strong validation checkpoints, explicit error paths, and safe-by-construction write constraints. Its weaknesses are redundancy — the hook procedure is duplicated almost verbatim before and after the main flow — and the absence of any progressive disclosure, with ~285 lines all inlined in SKILL.md.

Suggestions

Factor the duplicated extension-hook procedure (Pre-Execution Checks and Step 9) into a single shared section or a references file parameterized by the hook key (before_converge vs after_converge), cutting ~100 lines.

Move the detailed hook output templates and the extensions.yml parsing rules into a reference file (e.g., references/hooks.md) and keep only a one-line pointer in SKILL.md.

Make the code-scope keyword search concrete (e.g., suggest grep/ripgrep patterns or named directories) so Step 3's "keyword search for the concepts" is directly executable.

DimensionReasoningScore

Conciseness

Most sections are dense and directive, but the extension-hook procedure is duplicated nearly verbatim in Pre-Execution Checks and Step 9 (~100 of ~285 lines), plus an irrelevant tangent about escaping single quotes ("I'm Groot") — "mostly efficient but includes some unnecessary explanation or could be tightened" rather than anchor 2's pervasive padding.

3 / 5

Actionability

Highly concrete: an exact prerequisites command ("check-prerequisites.sh --json --require-spec --require-tasks --include-tasks"), absolute artifact paths, exact output templates, the ID format "T{M+1:03d}", and copy-ready checklist markdown. Minor gaps (e.g., "a keyword search for the concepts each requirement describes" specifies no concrete search command) keep it below anchor 5.

4 / 5

Workflow Clarity

The 9-step sequence has explicit validation checkpoints and feedback loops: prerequisites script with JSON parsing, a STOP with the named remediation command when artifacts are missing, an error path for unparseable extensions.yml, a pre-write findings summary, append-only guards, and explicit converged vs. tasks_appended branches with handoff guidance — matching anchor 5's clear sequence with explicit validation and error recovery.

5 / 5

Progressive Disclosure

The body is well-sectioned with ## / ### headers and numbered steps, but at ~285 lines with no external reference files, and with ~100 lines of duplicated hook machinery inlined, "content that should be separate is inline" — anchor 3 rather than 4, since the hook spec clearly belongs in a shared reference file.

3 / 5

Total

15

/

20

Passed

Description

58%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 communicates two concrete, artifact-anchored actions clearly and in third person, but it lacks any explicit "Use when..." trigger guidance and misses the skill's signature term (converge/convergence), which limits discoverability and completeness. It is distinctive within the spec-kit ecosystem but has minor overlap risk with the tasks/implement commands.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks to converge, check what's left unbuilt, or verify the implementation matches the spec after /speckit-implement has run."

Include the skill's signature terms ("converge", "convergence", "spec-kit") and natural user phrasings like "check remaining work" in the description to improve trigger-term coverage.

Briefly mention the outcome dimensions (findings classification by gap type and severity) to round out the "what" without adding fluff.

DimensionReasoningScore

Specificity

"Assess the current codebase against the feature's spec, plan, and tasks" and "append any remaining unbuilt work as new tasks to tasks.md" name concrete, artifact-specific actions with a clear output destination, but only two actions are listed — below anchor 5's comprehensive coverage yet more concrete than anchor 3's generic examples.

4 / 5

Completeness

The "what" is clear (assess codebase against artifacts, append remaining work), but there is no "Use when..." clause or equivalent explicit trigger guidance; per the judging guidelines a missing trigger clause caps completeness at 3, and the "when" is only weakly implied by "so implement can complete it".

3 / 5

Trigger Term Quality

Terms like "spec", "plan", "tasks", "tasks.md", and "codebase" are relevant, but the description omits the skill's own core term ("converge"/"convergence") and natural user phrases like "check what's left to build" — matching "some relevant keywords but missing common variations or synonyms" rather than anchor 4's good coverage.

3 / 5

Distinctiveness Conflict Risk

The spec/plan/tasks.md artifact framing carves a distinct niche, but "append any remaining unbuilt work as new tasks to tasks.md" overlaps with the closely related spec-kit tasks/implement commands, keeping it at anchor 4 ("mostly distinct; minor overlap risk") rather than 5.

4 / 5

Total

14

/

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

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

RihardsJ/software-factory

Path

apps/reference/.claude/skills/speckit-converge/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.