signals-scout-inbox-validation

Follow-up Signals scout for the inbox itself. After a deployment soak window, re-measures the problems behind recently resolved reports and files a report when a fix didn't hold, plus a gated escalation check on dismissed reports.

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./products/signals/skills/signals-scout-inbox-validation/SKILL.md

Quality

Content

88%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 a strong operational runbook: highly actionable with executable SQL and exact tool calls, a clearly sequenced multi-step workflow with explicit validation checkpoints and feedback loops, and good progressive disclosure that defers generic report mechanics to a clearly signaled reference. Its only real weakness is mild verbosity in the philosophical framing passages.

Dimension	Reasoning	Score
Conciseness	The body is information-dense and avoids explaining concepts Claude already knows — it supplies the specific SQL shape, the argMax/dedup gotchas, exact tool parameters, and verdict logic Claude could not infer — but a few philosophical framing passages ('Resolution-vs-reality is the signal-vs-noise discriminator', 'Honest unverfiability beats a fake probe', 'A team that merges a fix mentally closes the issue') are ethos statements that could be trimmed, matching the 4-anchor ('efficient; minor instances of over-explanation') better than the lean 5-anchor.	4 / 5
Actionability	Guidance is fully executable: exact tool calls with parameters (e.g. `inbox-reports-list {"status": "resolved", "ordering": "-updated_at", "limit": 20}`), complete runnable SQL (not pseudocode), concrete scratchpad key formats with example payloads, and a verdict table mapping observations to specific actions — copy-paste ready and covering the common cases, matching the 5-anchor.	5 / 5
Workflow Clarity	The run is clearly sequenced (Get oriented → Stage 1 enqueue → Stage 2 validate via a strongest-first probe ladder → verdict table → save memory → decide → close out) with explicit validation checkpoints (soak-window gate, denominator check, the verdict matrix), feedback loops (extend once then re-validate, two-passes-max), and checklists (verdict table, Disqualifiers); the batch nature is bounded by explicit caps (~5 enqueues, ~3 probes) and validation is present rather than missing, so the destructive/batch cap does not apply.	5 / 5
Progressive Disclosure	Structure is good with clear headers and an appropriately deferred one-level-deep reference for generic mechanics ('authoring-scouts → references/report-contract.md', repeatedly and clearly signaled, 'readable in-run via skill-file-get'), with the body keeping only inbox-validation specifics; it stops at 4 rather than 5 because the body is a long (~175 line) single file, the one reference is a cross-skill file rather than this skill's own bundle (no references/scripts/assets dirs exist), and the 'MCP tools' section partially restates tools already cited inline — minor organization gaps.	4 / 5
	Total	18 / 20 Passed

Description

46%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 conveys a clear, niche 'what' in third person, but leans heavily on internal jargon and omits any explicit 'Use when...' trigger guidance, leaving completeness and trigger-term quality as its weakest dimensions. Distinctiveness is strong thanks to the specialized inbox-validation framing.

Suggestions

Add an explicit 'Use when...' clause naming natural trigger phrases (e.g. 'Use when re-checking whether a recently resolved report's fix actually held, or when a dismissed report's underlying problem is escalating') to lift completeness and trigger-term quality.

Replace or gloss internal jargon ('soak window', 'resolved reports') with more natural, user-recognizable terms and include common synonyms so the description surfaces for the right requests.

Tighten the action verbs to be more concrete (e.g. 're-probes the error/log/metric entities behind a resolved report and compares post-fix rates to a pre-fix baseline') to move specificity from a jargon-laden high-level summary toward crisp enumerated actions.

Dimension	Reasoning	Score
Specificity	Names the domain ("Follow-up Signals scout for the inbox itself") and a few concrete actions ("re-measures the problems behind recently resolved reports", "files a report when a fix didn't hold", "gated escalation check on dismissed reports"), but the actions are couched in domain jargon and described at a fairly high level rather than as crisp operations, so it lands at the 'names domain and 1-2 concrete actions, not comprehensive' anchor rather than the 4-anchor's crisp enumerated verbs.	3 / 5
Completeness	The 'what' is clear (re-measure resolved-report fixes, file a report when a fix didn't hold, escalation check on dismissed reports), but there is no explicit 'Use when...' trigger clause and the 'when' is only weakly implied via 'After a deployment soak window'; per the rubric guideline a missing Use-when clause caps completeness at 3.	3 / 5
Trigger Term Quality	The keywords are internal jargon ("soak window", "resolved reports", "dismissed reports", "inbox") rather than natural phrases a user would say, and there is no "Use when..." clause supplying common variations; this matches the 2-anchor ('one or two generic keywords; missing the natural phrases users say') better than the 3-anchor, which expects relevant keywords with only missing synonyms.	2 / 5
Distinctiveness Conflict Risk	It carves a clear niche ('follow-up scout for the inbox itself', inbox validation of resolved/dismissed reports) with distinct triggers and low conflict risk, but the body explicitly references sibling Signals scouts ('the other scouts'), creating minor overlap risk with closely related scout skills, which fits the 4-anchor better than the 5-anchor.	4 / 5
	Total	12 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 3 missing	Warning

	Total	13 / 16 Passed

Repository: PostHog/posthog
Path: products/signals/skills/signals-scout-inbox-validation/SKILL.md
Commit: d293866

Reviewed: 5 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.