CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

rederive

This skill should be used when a contributor names a skill whose Lean contract is to be re-derived — a core protocol, a utility skill, or a contributor skill whose formal block is or becomes Lean — or invokes /rederive. Runs the per-skill flow: gather the pre-understanding, derive the skill's epistemic solution and fuse it with the contributor's horizon, then read the current contract against it, through sketch, edit and rehearsal, to the fusion of the contributor's horizon on the rehearsal result. Project-local contributor tooling.

Quality

67%

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

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/rederive/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%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 densely engineered formal Lean contract that is precise and internally complete, with real verification commands and a genuine one-level reference file. Judged as skill guidance, though, it buries executable steps inside formal notation, offers no plain operational entry point, and inlines nearly all detail in one monolithic block.

Suggestions

Add a short plain-language operational summary (the gate sequence: gather → derive → solution gate → read → sketch → edit → rehearse → hand-off) at the top of the body so the executable workflow can be followed without parsing the Lean state machine.

Move the long grounding strings for the heavier operations (`.read`, `.edit`, `.rehearse`) into `references/` files, keeping one-line summaries inline, to cut the monolithic block down and improve progressive disclosure.

Surface the verification commands (lean-contract.js check, lake build --wfail, lake test, static-checks) as an explicit checklist section rather than leaving them embedded inside the `.edit` string literal.

DimensionReasoningScore

Conciseness

The body contains no padding about concepts Claude already knows, and the prose (Definition, target paragraph) is extremely compressed. However, ~400 lines of dense Lean specification with very long, reference-laden doc strings (e.g. the multi-clause `grounding` strings for `.read` and `.edit`) could be materially tightened or split, matching anchor 3 rather than 4.

3 / 5

Actionability

Concrete, executable commands do appear — "git worktree add --detach <scratch> origin/main", "node .claude/skills/verify/scripts/lean-contract.js generate . && lake build --wfail", "git log -L", "/unfold decide" — but they are embedded inside dense Lean string literals within a formal state machine, and the operational path an executor should follow is never laid out as directly usable guidance. This sits between anchors 3 and 4, landing at 3 for the missing key execution details.

3 / 5

Workflow Clarity

The sequence is fully specified — the FLOW comment, `before` (gather→search→derive→consult; sketch→consult; edit→rehearse), `shows`, and the `run` state machine with gate checkpoints — and the edit operation carries explicit verification commands. But the workflow is expressed in formal Lean semantics and nested bracket conditions rather than a clear ordered procedure, leaving checkpoints implicit for the operational reader; anchor 3 fits better than 4.

3 / 5

Progressive Disclosure

The one bundle reference (`references/reading-checklist.md`) is real, clearly cited inline with a path ("Apply `references/reading-checklist.md` with file:line evidence"), and exactly one level deep. However, the body itself is a near-monolithic 400-line formal block with only two prose sections, and material such as the lengthy grounding strings could arguably live in reference files; anchor 3 (some structure, could be better organized) is the closest fit.

3 / 5

Total

12

/

20

Passed

Description

71%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 explicitly covers both when to use the skill and what it does, within a very distinctive niche, and lists the concrete stages of its flow. Its main weaknesses are jargon-dense phrasing that obscures the actions from a natural-language standpoint and limited coverage of trigger-term variations.

Suggestions

State the 'what' in plainer capability terms (e.g. "reads the current contract, sketches the change, edits it on a fork, and rehearses the edit") alongside the existing flow vocabulary.

Add natural trigger variations a contributor might actually say, such as "rederive", "re-derive", "update/rework a skill's Lean contract", so the skill surfaces for phrasings beyond the exact project vocabulary.

DimensionReasoningScore

Specificity

The description enumerates several concrete actions of the flow — "gather the pre-understanding, derive the skill's epistemic solution", "read the current contract against it, through sketch, edit and rehearsal" — giving a multi-step action list with only minor coverage gaps. It falls short of a 5 because the actions are named at a high level of abstraction (e.g. "fuse it with the contributor's horizon") rather than as fully concrete operations.

4 / 5

Completeness

Both parts are explicitly present: "This skill should be used when a contributor names a skill whose Lean contract is to be re-derived ... or invokes /rederive" answers 'when', and "Runs the per-skill flow: gather ... sketch, edit and rehearsal, to the fusion ... on the rehearsal result" answers 'what'. Not a 5 because the 'what' is stated in abstract, jargon-laden terms rather than plain concrete capability statements.

4 / 5

Trigger Term Quality

Relevant keywords are present — "re-derived", "invokes /rederive", "Lean contract", "core protocol" — but coverage of natural variations is thin and the phrasing is heavily project-internal jargon. It sits between anchor 3 (some relevant keywords, missing common variations) and anchor 4, settling at 3 because a contributor would likely phrase the need in ways ("redo", "rework", "update the contract") the description does not surface.

3 / 5

Distinctiveness Conflict Risk

The trigger conditions are highly specific — re-deriving a Lean formal contract for a named core protocol, utility skill, or contributor skill, or an explicit /rederive invocation — carving out a clear niche with minimal overlap risk against other skills.

5 / 5

Total

16

/

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

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

14

/

16

Passed

Repository

jongwony/epistemic-protocols

Path

.claude/skills/rederive/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.