root-cause-tracing

Use when errors occur deep in execution and you need to trace back to find the original trigger - systematically traces bugs backward through call stack, adding instrumentation when needed, to identify source of invalid data or incorrect behavior

Quality

72%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agency/plugins/nori/skills/root-cause-tracing/SKILL.md

Quality

Content

61%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 content is a genuinely useful, concrete debugging playbook with executable code and a clear 5-step tracing workflow. Its main weaknesses are a missing referenced bundle file (@find-polluter.sh is cited but not provided), some duplicated worked-example material, and a time-sensitive dated log line that pads tokens.

Suggestions

Either ship scripts/find-polluter.sh in the bundle or inline its contents; a referenced file that does not exist breaks progressive disclosure and leaves the 'find polluter' workflow unactionable.

Merge the 'Real Example: Empty projectDir' and 'Real-World Impact' sections — the trace chain, fix, and layered defenses are described twice; keep one concrete walkthrough.

Remove or relocate the dated 'From debugging session (2025-10-03)' log; time-sensitive incident notes don't belong in a reusable skill and add tokens without guidance value.

Dimension	Reasoning	Score
Conciseness	Mostly efficient, but the worked 'Real Example' / 'Real-World Impact' sections restate the same trace chain and layered fix twice, and a dated debugging-log line ('From debugging session (2025-10-03)') is time-sensitive padding that doesn't earn its tokens.	3 / 5
Actionability	Provides concrete, executable TypeScript for instrumentation, a runnable grep command, and a copy-paste bisection invocation; minor gap is that the referenced @find-polluter.sh script is not bundled, so the primary concrete tool is unavailable.	4 / 5
Workflow Clarity	The 5-step tracing process is clearly sequenced with a worked example, and the layered-fix section describes validation across four layers; it lacks an explicit validate→fix→retry feedback loop for the instrumentation/bisection batch operations, which keeps it just below 5.	4 / 5
Progressive Disclosure	The body references @find-polluter.sh as the key tool for finding polluting tests, but no ./scripts/, ./references/, or ./assets/ bundle exists and the file is missing — a referenced one-level path that cannot be navigated, plus a monolithic single-file body with no offloading.	2 / 5
	Total	13 / 20 Passed

Description

83%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 is strong: it opens with an explicit 'Use when...' trigger, names the domain and multiple concrete actions, and clearly states both what the skill does and when to invoke it. It could be improved with a few more natural synonyms ("root cause", "stack trace") but is otherwise well above average.

Suggestions

Add natural synonyms a user might actually say, e.g. 'root cause', 'stack trace', or 'where did this bad value come from', to broaden trigger coverage.

Tighten the description slightly — 'systematically traces bugs backward through call stack, adding instrumentation when needed, to identify source of invalid data or incorrect behavior' repeats the 'trace back' idea twice and could be trimmed.

Confirm third-person voice is consistent throughout; the current phrasing is third-person, which is good.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions — "systematically traces bugs backward through call stack", "adding instrumentation", "identify source of invalid data" — with only minor gaps in coverage (no explicit synonym coverage).	4 / 5
Completeness	Explicitly answers both halves: what ("systematically traces bugs backward through call stack, adding instrumentation... to identify source") and when ("Use when errors occur deep in execution and you need to trace back").	5 / 5
Trigger Term Quality	"errors occur deep in execution", "trace back to find the original trigger", "call stack" are natural phrases a user would say, but it lacks synonyms like "stack trace", "root cause", or "where did this bug come from".	4 / 5
Distinctiveness Conflict Risk	The deep-stack backward-tracing niche is fairly distinct from generic debugging skills, though "trace bugs" could loosely overlap with general debugging skills.	4 / 5
	Total	17 / 20 Passed

Validation

100%

Warnings & errors only

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: microsoft/FluidFramework
Path: .agency/plugins/nori/skills/root-cause-tracing/SKILL.md
Commit: f02e424

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

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.