doubt-driven-development

Subjects every non-trivial decision to a fresh-context adversarial review before it stands. Use when you want every assumption cross-examined before proceeding, when stress-testing a plan for hidden failure modes, when correctness matters more than speed, when working in unfamiliar code, when stakes are high (production auth, security-sensitive logic, a high-stakes migration, irreversible operations), or any time a confident output would be cheaper to verify now than to debug later.

Quality

74%

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

70%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 strong process skill with an exceptionally clear five-step workflow, explicit feedback loops, and mostly copy-paste-ready prompts. It is held back by repetition in the cross-model escalation material and by inlining substantial content that would be better split into reference files.

Suggestions

Extract the cross-model escalation subsection (PATH check, CLI verification, invocation examples, failure handling) into a dedicated reference file and keep a short 'always offer; see references/cross-model.md' summary in the main body — this would improve both conciseness and progressive disclosure.

De-duplicate the repeated cross-model framing: 'Skipping is fine; silent skipping is not' and 'the user decides, the agent surfaces the choice' each appear multiple times across the escalation section, the rationalizations table, and Red Flags; state each once and cross-reference.

The rationalizations table partially re-states guidance already given in Red Flags and the Verification checklist; merge overlapping rows (e.g., the cross-model and reviewer-deference rows) to cut token cost without losing coverage.

Dimension	Reasoning	Score
Conciseness	The guidance is novel and mostly earns its tokens (adversarial prompt, reconcile precedence, stop conditions), but there is noticeable repetition: "Skipping is fine; silent skipping is not" appears twice, and "the user decides / the agent surfaces the choice" is restated three times across the cross-model section, the rationalizations table, and Red Flags. Mostly efficient but could be tightened — anchor 3.	3 / 5
Actionability	Provides a verbatim copy-paste adversarial prompt, CLAIM/EXTRACT templates, a concrete four-class reconcile precedence order, and example CLI invocations with stdin-piping guidance. Minor gaps keep it from a 5: the CLI examples self-caveat that flags must be verified, and the reviewer roster is delegated to an `agents/` directory that is not part of this skill's bundle.	4 / 5
Workflow Clarity	Five clearly sequenced steps (CLAIM, EXTRACT, DOUBT, RECONCILE, STOP) with a copyable checklist, explicit feedback loops (re-loop on actionable findings, 3-cycle bound with user escalation), a classification procedure for findings, and a final verification checklist — matching the anchor for clear sequence with explicit validation and error-recovery loops.	5 / 5
Progressive Disclosure	The ~240-line body is well-sectioned but monolithic: the ~50-line cross-model escalation subsection and the rationalizations table are prime candidates for separate reference files. The only file references (`../../references/orchestration-patterns.md` and `agents/`) point outside the skill directory and no bundle files exist in this skill to back them, so navigation to deeper material is not actually available — anchor 3.	3 / 5
	Total	15 / 20 Passed

Description

78%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 well-constructed description with an explicit what, rich and natural trigger phrasing, and concrete high-stakes examples. Its main weaknesses are that it conveys a single core action rather than a range of capabilities, and it does not distinguish itself from generic review skills.

Dimension	Reasoning	Score
Specificity	"Subjects every non-trivial decision to a fresh-context adversarial review before it stands" names the domain and one concrete core action, but the description lists no additional distinct actions — everything after the first sentence is trigger phrasing. This matches the 'domain and 1-2 concrete actions' anchor rather than anchor 4, which requires several specific actions.	3 / 5
Completeness	Explicitly answers both questions: the what is "Subjects every non-trivial decision to a fresh-context adversarial review before it stands", and an explicit "Use when" clause lists concrete trigger scenarios. This matches the anchor-5 example pattern (clear what + explicit when with concrete triggers).	5 / 5
Trigger Term Quality	Strong natural-phrase coverage: "stress-testing a plan for hidden failure modes", "correctness matters more than speed", "working in unfamiliar code", and concrete stake examples like "production auth, security-sensitive logic, a high-stakes migration, irreversible operations". Not a 5 because common user phrasings like "second opinion", "red-team my plan", or "challenge my assumptions" are absent.	4 / 5
Distinctiveness Conflict Risk	"Fresh-context adversarial review before it stands" carves a distinct in-flight verification niche with mostly distinct triggers. Minor overlap risk remains with general review/code-review skills, which the description does not disambiguate from — anchor 4, not 5.	4 / 5
	Total	16 / 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: 2686b62

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.