CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

bridge

Capture a cross-project connection the moment it surfaces, such as a shared pattern, a person carrying context between projects, a story that lends another project credibility, a feedback loop. Use when one project feeds, reuses, reinforces another: 'connects to', 'feeds into', 'same pattern as'. Writes one YAML file per bridge to a bridges directory, with list, map and stats subcommands.

Quality

90%

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

78%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.

A well-structured skill body: an explicit 8-step procedure with exact failure messages and validation checks, lean non-pedantic prose, and clean one-level reference files verified to exist. The only deductions are minor: duplicated rules between Principles and Anti-Patterns, an inline bash snippet referencing undefined variables with an inconsistent path, and no recovery step when the parse check fails.

Suggestions

Define '$dir' and '$file' in the inline bash snippet (or remove the duplicated snippet and rely solely on storage-and-schema.md) and use the same '${PRAXIS_DIR}/thinking/bridges' path in both places so the snippet is copy-paste ready.

Add an explicit recovery step after the yq parse check, e.g. 'If parsing fails, re-quote the description/context and rewrite the file before answering', to close the workflow's only feedback-loop gap.

Trim the overlap between Principles and Anti-Patterns — 'Store the user's words, not yours' and 'Choose an honest default over a confident guess' restate the NEVER-reformulate and NEVER-guess-a-rare-archetype bullets; keep one formulation of each rule.

DimensionReasoningScore

Conciseness

The body is lean, assumes Claude's competence, and explains no general concepts, but there is minor duplication that could be trimmed: 'Store the user's words, not yours' restates the 'NEVER reformulate the user's description' anti-pattern, 'Choose an honest default over a confident guess' restates the 'NEVER guess a rare archetype' bullet, and the inline sequence-number bash duplicates the snippet in storage-and-schema.md. This is the efficient-with-minor-trimming tier, not the every-token-earns-its-place tier.

4 / 5

Actionability

Guidance is mostly executable: exact input grammar ('/bridge <src> → <tgt>: <desc>' with '[type]'), exact response template, ordered keyword lists for strength detection, and a bash snippet for the sequence number and parse check. Minor gaps keep it below fully copy-paste ready: the inline bash uses '$dir' and '$file' without defining them in the body, and it hardcodes './thinking/bridges/' while the storage reference uses '${PRAXIS_DIR}/thinking/bridges'.

4 / 5

Workflow Clarity

The 8-step procedure is clearly sequenced with most checkpoints present: stop-on-missing-config with an exact warning message, write-nothing-on-unresolved-alias, duplicate check before writing, collision-safe numbering (highest number plus one), and a post-write yq parse confirmation. It falls short of the top anchor because the parse check has no explicit recovery path (no 'if parse fails, fix and re-write' step), leaving a minor validation gap.

4 / 5

Progressive Disclosure

The body is a genuine overview (principles, when/when not, procedure, anti-patterns) and the four detail files — archetype detection, storage/schema, list/map/stats, worked examples — all exist, are one level deep with no nested references, and are well signaled both in the References section with descriptions and inline at the point of use (steps 4, 7, and Commands). Content is appropriately split and easy to navigate.

5 / 5

Total

17

/

20

Passed

Description

100%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.

An exemplary description: third-person voice, concrete actions with storage format and subcommands, an explicit 'Use when' clause with quoted natural trigger phrases, and a distinctive niche. It reads like the reference good examples and shares none of the bad examples' vagueness.

DimensionReasoningScore

Specificity

Multiple concrete actions are listed comprehensively: 'Capture a cross-project connection', 'Writes one YAML file per bridge to a bridges directory', plus 'list, map and stats subcommands', with four concrete instantiations (shared pattern, person carrying context, story lending credibility, feedback loop). This matches the anchor for multiple specific concrete actions with comprehensive coverage; it is above the 'minor gaps' level because the storage format, location, and read-back subcommands are all named.

5 / 5

Completeness

Both questions are explicitly answered: what ('Capture a cross-project connection... Writes one YAML file per bridge to a bridges directory, with list, map and stats subcommands') and when ('Use when one project feeds, reuses, reinforces another: "connects to", "feeds into", "same pattern as"') with concrete trigger phrases — the exact top-anchor pattern.

5 / 5

Trigger Term Quality

Natural trigger phrasings users would actually say are quoted verbatim ('connects to', 'feeds into', 'same pattern as') alongside synonyms ('feeds, reuses, reinforces') and domain signals ('shared pattern', 'carrying context', 'feedback loop'). Coverage is comprehensive for this domain; file extensions are not applicable, so nothing natural is missing.

5 / 5

Distinctiveness Conflict Risk

It carves a clear niche (cross-project connection capture with its own vocabulary — bridges, archetypes, YAML capture files) and its triggers ('connects to' another project, 'same pattern as') would not fire for unrelated skills. Minimal conflict risk; clearly distinct from generic note-taking or documentation skills.

5 / 5

Total

20

/

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

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository

pantheon-org/tekhne

Path

skills/software-engineering/bridge/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.