AC long-research
[BETA] Deep research that actually reads pages instead of summarizing search results. Tell it how long to research (10 min, 2 hours, all night) and it works the full duration — searching, reading every result, following leads, cracking forums, cross-verifying findings, and writing progressively to a research file. Tree-style exploration: each page read spawns new searches, like a human researcher. Enforced read-to-search ratio prevents shallow search-spamming. Wall-clock time commitment — it won't finish early. Self-audit gate blocks delivery until quality checks pass. Works with web_search, web_fetch, and browser-use for JS-heavy sites.
[BETA] Deep research that actually reads pages instead of summarizing search results.
As a process C 58/100 · Has gaps — weak spots: inputs and preconditions, execution cost, running it twice
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- 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: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 9707 tokens (recommended < 5000); move details to references/ - note
frontmatter-keyunknown frontmatter key "dependencies"
Process rating: all ten parameters 58/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 10 mutating operations with no state check
- 40Execution cost. Instruction body is 9707 tokens: crowds the task out of the window
- 50When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 60Steps. 208 steps, 5 vague phrases
- 60Result and completion. Output format stated, no completion criterion
- 100Failures and branches. 6 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 15 top-level sections: this looks like several domains in one skill
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)
- -267 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +3Description length 645: enough signal without eating the budget
- +4Structure: 48 headings
- +3Step-by-step instructions: 208 items
- +3Output format is stated explicitly
- +4Has examples (15 code blocks)
- +4Reference files are cited in the instructions (2 of 3)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 78.