Parallel FindAll
PassAudited by ClawScan on May 7, 2026.
Overview
The skill is coherent and purpose-aligned, but users should notice that it relies on an authenticated external Parallel CLI account and can optionally hand large results to a helper agent.
This skill appears safe to install if you intend to use Parallel FindAll. Before using it, make sure `parallel-cli` is installed from Parallel's official docs, confirm which Parallel account is authenticated, and approve large or pro-tier runs that may use more quota. For large result sets, be aware that the skill may save JSON to /tmp and ask a helper agent to summarize it.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may run Parallel FindAll commands under the user's environment and create or manage remote discovery jobs.
The skill instructs the agent to use a local CLI to create and manage remote FindAll jobs. This is expected for the stated purpose, but it is still provider tool use that may consume account quota or change remote job state.
parallel-cli findall run "<objective>" [options] parallel-cli findall status frun_xxx --json parallel-cli findall poll frun_xxx --json parallel-cli findall result frun_xxx --json parallel-cli findall extend frun_xxx <n> --json parallel-cli findall cancel frun_xxx
Use reasonable match limits, be cautious with pro-tier or large runs, and confirm before running jobs that may be costly or high-volume.
FindAll requests will run through the user's authenticated Parallel account and may expose the discovery objective to Parallel or use that account's quota.
The skill depends on an already-authenticated Parallel CLI profile, even though registry metadata does not declare a primary credential. This is purpose-aligned, but users should know which account is being used.
Requires `parallel-cli` (installed and authenticated). If `parallel-cli --version` fails, or if a later command fails with an authentication error, tell the user to see https://docs.parallel.ai/integrations/cli and stop.
Verify the active Parallel account and credentials before use, and avoid sending sensitive objectives unless the user is comfortable with Parallel processing them.
The safety of the actual CLI depends on what the user has installed locally.
The provided skill does not install or bundle code, but its instructions require a separately installed `parallel-cli`. The external CLI itself is outside the supplied artifact set and was not reviewed here.
No install spec — this is an instruction-only skill.
Install or update `parallel-cli` only from the official Parallel documentation and verify the binary before authenticating.
Large result sets may be written to a local temporary file and processed by a spawned helper agent.
For large outputs, the skill suggests saving results to a temporary file and passing that file to another agent. The task is bounded to a specific result file, but it is still an inter-agent data flow.
Then spawn a sub-agent:
{"tool": "sessions_spawn", "task": "Read /tmp/findall-<topic>.json and produce a deduplicated table grouped by category.", "label": "findall-summary"}Use this helper flow only when needed, avoid including sensitive data in saved result files, and delete temporary files after use.