CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ctx-search

Search context-mode's persistent FTS5 knowledge base for previously indexed local project content, documentation, or session memory. Trigger: /context-mode:ctx-search

Quality

83%

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

100%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 exemplary lean, single-purpose skill: concrete executable examples on both the MCP and CLI paths, explicit fallback and empty-index failure handling, and no wasted tokens. No changes needed for a skill of this scope.

DimensionReasoningScore

Conciseness

The body is 36 lines of pure directive and example content with zero concept explanation — the single prose sentence "Search indexed content without rereading raw sources into conversation context" justifies itself by stating the purpose. Every token earns its place, matching the lean anchor 5; there is no padding to trim for a 4.

5 / 5

Actionability

Both execution paths get concrete, copy-paste-ready examples with realistic values: the MCP call `ctx_search({source: "project:<name>", queries: ["authentication middleware", "token refresh"], limit: 5})` and the CLI `context-mode search "authentication middleware" --source project:<name> --limit 5`. Guidance like "Use short, specific queries of two to four technical terms" is specific and measurable, matching anchor 5 rather than 4 because the common cases are fully covered.

5 / 5

Workflow Clarity

The numbered flow covers the primary path (steps 1–4: prefer MCP, batch queries, scope by source, query style), an explicit fallback (step 5: "If MCP tools are unavailable, fall back to the CLI"), and failure recovery (step 6: empty index → remediation via "/context-mode:ctx-index or context-mode index <path>"). This is a single-action, non-destructive skill where the simple-skill exception applies, and the fallback plus failure handling go beyond the unambiguous-single-action bar.

5 / 5

Progressive Disclosure

The skill is under 50 lines, self-contained, and needs no external reference files (none exist in the bundle), with clean section organization (title, purpose line, Instructions with embedded examples). Per the rubric guideline, skills under 50 lines with no need for external references score 5 with well-organized sections, which this body satisfies.

5 / 5

Total

20

/

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 gives a clear, technically specific statement of what the skill does and defines a distinct niche, but its trigger guidance is limited to a slash command. Adding a natural-language "Use when…" clause with user-facing trigger phrases would substantially improve completeness and trigger term quality.

Suggestions

Add an explicit "Use when…" clause, e.g. "Use when the user wants to search previously indexed content, look up past session memory, or query indexed project documentation without rereading source files."

Include natural trigger phrases and synonyms users would actually say (e.g. "search the index", "find indexed docs", "look up session memory") rather than relying solely on the slash-command trigger.

Optionally mention the complementary indexing skill by name in the when-clause so users landing here with an empty index immediately know the companion workflow.

DimensionReasoningScore

Specificity

"Search context-mode's persistent FTS5 knowledge base for previously indexed local project content, documentation, or session memory" names one action but grounds it in a concrete technology (persistent FTS5 index) and enumerates three searchable content types. It is not 5 because only a single action is listed rather than multiple concrete actions; it exceeds 3 because the action is fully and specifically scoped rather than generically described.

4 / 5

Completeness

The "what" is clear (search a persistent FTS5 knowledge base of indexed content), but the "when" is only weakly implied via a slash-command trigger with no "Use when…" clause or equivalent usage conditions. Per the judging guideline, a missing explicit "when" caps completeness at 3; it cannot be 4 because the trigger names how to invoke, not when the skill should be used.

3 / 5

Trigger Term Quality

Relevant keywords exist ("search", "documentation", "session memory", "project content") but the only trigger guidance is the slash command "Trigger: /context-mode:ctx-search", which is an invocation path rather than natural language a user would say. Common variations and synonyms are missing, matching anchor 3; it is not 4 because no natural trigger phrases are given, and not 2 because several genuinely relevant keywords are present.

3 / 5

Distinctiveness Conflict Risk

"context-mode's persistent FTS5 knowledge base" carves out a distinct niche tied to a specific subsystem, keeping conflict risk low. It is not 5 because the absence of distinctive natural-language triggers leaves minor overlap risk with generic search or documentation-lookup requests; it clearly exceeds 3 since the domain is specific rather than broadly overlapping.

4 / 5

Total

14

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

mksglu/context-mode

Path

skills/ctx-search/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.