bounded-researcher

v0.1.0

Bounded evidence-first research workflow for software agents. Use when an agent should reduce uncertainty, localize issues, validate outputs, or summarize ev...

0· 184·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The name/description (bounded, evidence-first research) matches the SKILL.md workflow and outputs; nothing requested (no env vars, binaries, or installs) is out of scope for that purpose.
Instruction Scope
Instructions are narrowly scoped to triage/localize/validate/summarize tasks and explicitly limit scope (load minimal context, escalate when out-of-scope). They do not instruct broad file reads, external endpoints, or credential access.
Install Mechanism
No install spec or code files are included; the skill is instruction-only so nothing is written to disk or downloaded during install.
Credentials
The skill declares no environment variables, credentials, or config paths; there are no disproportionate or unexplained secret requests.
Persistence & Privilege
always is false and autonomous invocation defaults are standard; the skill does not request permanent presence or to modify other skills or system-wide settings.
Assessment
This skill is a set of behavioral instructions (no code, no installs, no credentials) and appears coherent with its stated purpose. Before installing, consider: (1) enforcement — these are guidelines the agent will try to follow but your platform controls actual file and network access, so restrict the agent's runtime permissions if you want to prevent any file reads or external calls; (2) supervision — pair this skill with a coordinator/supervisor as intended to avoid scope creep; (3) testing — run it in a sandboxed environment first to confirm outputs and that it escalates rather than makes implementation changes. If you require strong enforcement of the “do not touch” rules, ensure your agent runtime policy limits file and network access accordingly.

Like a lobster shell, security has layers — review code before you run it.

latestvk9710rb2e0cnxsc3022e2g4yvd82q8vp

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments