CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

para-vault

Use this skill when the user asks to "file this note into PARA", "process vault inbox", "organize my PARA vault", "reconcile vault documents", "propose note revision", "archive this project", "reactivate archived note", "check vault navigation", "repair broken MOC links", or "import calendar/email records into vault". Covers the full PARA knowledge lifecycle (Projects, Areas, Resources, Archives). Do NOT use this skill for raw filesystem moves outside a PARA vault (use `manage-file-operations`) or drafting user communications (use `write-like-me`).

Quality

75%

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 ./skills/para-vault/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

53%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-written overview in isolation — clean routing table, tight invariants, and a genuine verification checklist — but it is wired to a bundle that doesn't match it: all five workflow runbook links are dead, and most of the actual reference material is orphaned with no entry point from SKILL.md. The skill functions as a directory to a building whose rooms are all locked and unlisted.

Suggestions

Fix the five dead runbook links: either create the workflows/*.md files or re-point the triage matrix rows at the existing reference files that cover them (e.g. workflows/file.md → references/classification.md + folder-conventions.md; workflows/reconcile.md → references/revisions.md; workflows/archive.md → references/lifecycle.md; workflows/navigate.md → references/navigation.md; workflows/import-records.md → references/source-records.md).

Link the remaining orphaned references (vault-conventions.md, and the ones not covered above) from an explicit 'References' section so every bundle file is reachable from SKILL.md in one hop.

Collapse the duplicated routing — the ASCII lifecycle tree and the triage table convey the same five workflow mappings; keep the table (it adds the intent-matching and output focus) and trim the tree or reduce it to the three phase headings.

DimensionReasoningScore

Conciseness

The body is compact (~55 lines), teaches nothing Claude already knows, and uses dense invariants ("evaluate in strict order: Projects (active, finite) → Areas → Resources → Archives") and specific field names ("status, completed, archived-from"). It matches the 'efficient; minor instances that could be trimmed' anchor: the lifecycle ASCII tree and the triage table both map the same five workflows, and the top scope directive partially repeats the frontmatter's negative boundaries. Not 3 because there is no real over-explanation, only this duplicated routing.

4 / 5

Actionability

The triage matrix and invariants are concrete, but the primary executable paths — every "Target Runbook" link ([`workflows/file.md`](workflows/file.md) through [`workflows/import-records.md`](workflows/import-records.md)) — point to files that do not exist in the bundle, so Claude cannot actually execute the routed task. This matches the 'some concrete guidance but incomplete' anchor. Not 4 ('mostly executable with minor gaps'): the gap is not minor — all five runbooks are dead links, and the real detail files that do exist (e.g. revisions.md, source-records.md) are never surfaced.

3 / 5

Workflow Clarity

Routing-level sequencing is clear (three lifecycle phases, intent-to-runbook table) and the Verification section supplies explicit checkpoints for a destructive/batch skill, so the validation cap is not triggered. However, the actual step-by-step workflows live exclusively behind the five nonexistent workflows/*.md links, so the detailed sequences and their per-task checkpoints are unreachable — matching 'steps listed but gaps; checkpoints missing or implicit' at the level where work is actually done. Not 4: a minor validation gap would still leave executable runbooks behind the links.

3 / 5

Progressive Disclosure

Scored against the actual bundle per the guidelines: the body links to 5 workflows/*.md files that do not exist, while 7 of the 8 real reference files (classification, folder-conventions, lifecycle, navigation, revisions, source-records, vault-conventions) are never referenced from SKILL.md — they are buried and undiscoverable. Only references/frontmatter-schemas.md is both real and linked. This matches the 'references are buried' anchor. Not 3: the problem is not inlining — it is that the declared structure does not match the bundle and most detail content is unreachable. Not 1 only because the SKILL.md body itself is cleanly sectioned and one-level linking is attempted; vault-conventions.md's hop to ../../manage-file-operations/core.md is a cross-skill dependency rather than gratuitous nesting.

2 / 5

Total

12

/

20

Passed

Description

86%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 strong description: explicit, natural trigger phrases covering the full PARA lifecycle, third-person-compatible capability framing, and exemplary negative boundaries naming the adjacent skills. The only weakness is that the 'what' is expressed as a domain statement with actions implicit in the trigger quotes rather than as a direct capability list.

DimensionReasoningScore

Specificity

Ten concrete operations are enumerated via quoted phrases ("file this note into PARA", "reconcile vault documents", "archive this project", "repair broken MOC links", "import calendar/email records"), matching the 'several specific actions with minor gaps' anchor. It falls short of 5 because the direct capability statement is domain-level ("Covers the full PARA knowledge lifecycle (Projects, Areas, Resources, Archives)") rather than a crisp action list; the actions live inside the trigger phrases instead. It is well above 3 ('1-2 concrete actions') given the breadth of concrete operations named.

4 / 5

Completeness

'When' is exemplary — an explicit "Use this skill when the user asks to..." clause with concrete trigger phrases — and 'what' is answered by "Covers the full PARA knowledge lifecycle (Projects, Areas, Resources, Archives)" plus the actions embedded in the triggers. It sits below 5 because the 'what' statement itself is lifecycle/domain vocabulary rather than an explicit capability list; a reader must infer the operations from the trigger quotes. Well above 4's typical weakness (an under-specified 'when'), which this description does not have.

4 / 5

Trigger Term Quality

The description leads with ten natural, verbatim user phrasings spanning the skill's full range ("organize my PARA vault", "process vault inbox", "reactivate archived note", "check vault navigation") including synonyms (file/organize, archive/reactivate), matching the 'comprehensive coverage of natural terms including synonyms' anchor. File extensions are absent but inapplicable — PARA vault operations have no distinctive file types. Not 4: coverage leaves no obvious natural variation missing.

5 / 5

Distinctiveness Conflict Risk

The description carves a clear niche (PARA vault lifecycle) and adds explicit negative routing — "Do NOT use this skill for raw filesystem moves outside a PARA vault (use `manage-file-operations`) or drafting user communications (use `write-like-me`)" — disambiguating the two most confusable sibling skills by name. This matches the 'clear niche with distinct triggers; minimal conflict risk' anchor exactly.

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: 5 missing

Warning

Total

15

/

16

Passed

Repository

NyssaAI/agent-skills

Path

skills/para-vault/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.