doubt-driven-development

Subjects every non-trivial decision to a fresh-context adversarial review before it stands. Use when correctness matters more than speed, when working in unfamiliar code, when stakes are high (production, security-sensitive logic, irreversible operations), or any time a confident output would be cheaper to verify now than to debug later.

Quality

78%

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 ./skills/doubt-driven-development/SKILL.md

Quality

Content

73%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 well-structured, highly actionable process skill with an excellent sequenced workflow and validation checkpoints. Its weakness is conciseness: several load-bearing rules are repeated across multiple sections, inflating the token budget without adding new guidance.

Suggestions

Consolidate the repeated rules (do-not-pass-CLAIM, adversarial-not-validating, offer-cross-model-every-cycle) into their primary step and have the Rationalizations table and Red Flags reference them rather than re-stating them in full.

Move the Common Rationalizations table and/or the Cross-model escalation detail into a reference file under references/ so SKILL.md stays a lean overview, improving both conciseness and progressive_disclosure.

Trim the Overview's contrast with /review, which is already made in the Rationalizations row and the 'Interaction with Other Skills' section.

Dimension	Reasoning	Score
Conciseness	The body avoids explaining concepts Claude already knows, but restates load-bearing points ('do not pass the CLAIM', 'adversarial not validating', 'cross-model offer every cycle') across Overview, Step text, the Rationalizations table, Red Flags, and Verification, giving it more than minor over-explanation.	3 / 5
Actionability	It provides copy-paste-ready adversarial prompt text, a RECONCILE classification with explicit precedence, concrete codex/gemini bash invocations, and a verification checklist, with only placeholders and tool-variant uncertainty keeping it from a 5.	4 / 5
Workflow Clarity	A clear 5-step CLAIM/EXTRACT/DOUBT/RECONCILE/STOP sequence is paired with a copyable checklist, explicit feedback loops (classify-then-re-loop, 3-cycle bound, decompose-if-too-big), escalation rules, and a final verification checklist, matching anchor 5.	5 / 5
Progressive Disclosure	No bundle files exist, so all content lives in one well-sectioned SKILL.md; references to orchestration-patterns.md and agents/ are clearly signaled and one level deep, but the Rationalizations/Red Flags material could be split out, keeping it at 4 rather than 5.	4 / 5
	Total	16 / 20 Passed

Description

82%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 well-constructed: third-person voice, explicit what-and-when, distinct niche. Its main limitation is naming a single core action rather than enumerating several concrete capabilities, which holds specificity at the mid-range.

Dimension	Reasoning	Score
Specificity	It names one concrete action concretely — 'Subjects every non-trivial decision to a fresh-context adversarial review before it stands' — but does not list several specific actions, so it fits anchor 3 rather than the multi-action anchor 4.	3 / 5
Completeness	It clearly answers 'what' (subjects non-trivial decisions to a fresh-context adversarial review) and 'when' with an explicit 'Use when' clause listing multiple concrete trigger phrases, matching the anchor 5 example.	5 / 5
Trigger Term Quality	Natural situational triggers like 'correctness matters more than speed', 'unfamiliar code', 'stakes are high (production, security-sensitive logic, irreversible operations)' give good coverage a user would voice, short of comprehensive synonym-level coverage at 5.	4 / 5
Distinctiveness Conflict Risk	The 'fresh-context adversarial review before it stands' niche is distinct and explicitly contrasted with /review, giving it a clear niche with minimal conflict risk per anchor 5.	5 / 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: addyosmani/agent-skills
Path: skills/doubt-driven-development/SKILL.md
Commit: df1edb2

Reviewed: about 18 hours 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.