AC descript
The Descript craft skill — edit talk content (podcasts, interviews, talking-head video) by editing the transcript instead of the timeline. Use when someone wants to edit in Descript, edit a podcast or interview, remove filler words/silences, clean up audio (Studio Sound), fix a flubbed word without re-recording (Overdub), auto-cut between speakers, turn one recording into clips + show notes + chapters, or asks about Descript's plans, credits, or Underlord. Uses the WORDS framework plus the interview rule: concision yes, meaning-flips never. Reads the recording's content skill + brand-profile/voice-builder first. The agent plans the edit (API/MCP where connected); the HUMAN verifies by ear and approves; WoopSocial publishes the exports. Overdub is consent-verified own-voice-only; tiers/credits are verified in-app. Distinct from capcut (visual short-form), captions-and-clipping/opus-clip (clip selection at scale), ai-voiceover (dedicated TTS), and podcast-and-audiograms (the strategy).
The Descript craft skill — edit talk content (podcasts, interviews, talking-head video) by editing the transcript instead of the timeline.
As a process C 56/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 6. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 56/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 8 mutating operations with no state check
- 40Result and completion. Does not say what the result is
- 100Tools and files. No external tools needed
- 100Steps. 9 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1774 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- high The skill tells the model to perform an irreversible action with no human approval
- low No test case covers injection arriving through data
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 998: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +4No input/output examples
- +1No license
- +2Single-language instructions
- +4Structure: 9 headings
- +3Step-by-step instructions: 9 items
- +4Reference files are cited in the instructions (4 of 4)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 81.