CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

latency-instrumentation

Surgically injects trace macros (TRACE_EVENT) into Chromium C++ files to expose "black box" latency gaps, resolves headers, and compiles the browser using a self-healing loop. Strictly no performance optimizations or refactoring.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

—

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./agents/skills/latency-instrumentation/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

81%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, actionable workflow that shines in sequencing and error recovery: the self-healing compile loop with bounded retries, checkpoints, and a guarded revert is exemplary. Minor redundancy, a typo, and undefined environment-specific tool names keep conciseness and actionability just short of excellent.

Suggestions

Deduplicate the revert rule and the no-optimizations warning — state each once (e.g., fold the Sandbox & Safety revert bullet into Step 4's revert step) to tighten conciseness.

Fix the typo "file content modificatio tools" and briefly clarify what "codesearch tool" / "view_file" refer to, or drop them in favor of standard search/read commands.

Consider moving the failure-repair heuristics (Step 4.3 error taxonomy) or the output contract into a short reference section/file if the skill grows, keeping the core loop lean.

DimensionReasoningScore

Conciseness

The body is efficient and assumes C++/git competence, but has minor trimmable redundancy: the no-optimizations constraint is stated in both the opening section and the CRITICAL DESIGN CONSTRAINT section, and the `git reset --hard HEAD` revert rule appears in both the safety section and Step 4, plus a typo ("modificatio tools").

4 / 5

Actionability

Provides mostly executable guidance — concrete commands for search (`rg -n --type cpp`), build (`autoninja -C out/Default chrome`), formatting (`git cl format`), a specific commit message, and a JSON output contract — with minor gaps: the TRACE_EVENT snippet uses generic placeholders and tool names like "codesearch tool" and "view_file" are referenced without definition.

4 / 5

Workflow Clarity

Clear four-step sequence with explicit validation (compile check) and a full feedback loop: parse compiler stderr, apply targeted repair, re-run autoninja, bounded to 3 attempts, with commit checkpoints and a guarded `git reset --hard HEAD` revert on hard failure — matching the anchor for explicit validation steps and error-recovery loops.

5 / 5

Progressive Disclosure

The skill is a single well-organized file with clear sections, numbered steps, and visual separators, and no bundle files exist so nothing is mis-placed; however, at roughly 120 lines of body with all content inline and no external references, it sits just below the fully-optimized structure anchor.

4 / 5

Total

17

/

20

Passed

Description

70%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 highly specific, distinctive description written in third person that clearly conveys what the skill does and explicitly fences out adjacent use cases. Its main weakness is the absence of any "Use when..." trigger guidance, which caps completeness and leaves invocation conditions implicit.

Suggestions

Add an explicit trigger clause, e.g. "Use when instrumenting Chromium main-thread latency gaps with TRACE_EVENT macros or when asked to add tracing to uninstrumented C++ code."

Include common user-facing synonyms such as "instrument", "tracing", and "profiling" to broaden natural keyword coverage.

Optionally mention the locate-and-report steps (finding target files, returning a compile-status JSON) so the action list is comprehensive.

DimensionReasoningScore

Specificity

Lists several concrete, domain-specific actions — "injects trace macros (TRACE_EVENT) into Chromium C++ files", "resolves headers", "compiles the browser using a self-healing loop" — but stops short of comprehensive coverage (e.g., locating target files and reporting results are unstated), matching the anchor for several specific actions with minor gaps.

4 / 5

Completeness

The "what" is clearly and concretely stated, but there is no "Use when..." clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Includes strong natural keywords for the niche — "trace macros", "TRACE_EVENT", "Chromium C++", "latency gaps", "black box", "compiles the browser" — though common variations a user might say such as "instrument", "tracing", or "profiling" are missing, placing it just below comprehensive coverage.

4 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (Chromium TRACE_EVENT injection) with distinct triggers and an explicit boundary ("Strictly no performance optimizations or refactoring") that further separates it from optimization or general C++ skills, so conflict risk is minimal.

5 / 5

Total

16

/

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

chromium/chromium

Path

agents/skills/latency-instrumentation/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.