skilless.ai research

Data Community

Skill notes

Routing Table

Trigger words · Tool · Purpose search, find, look up · search.py · Web, docs, news lookup read, fetch, get page · web.py · Extract page text download, extract, subtitles, transcript · youtube.py / yt-dlp · Video subtitles and metadata convert, compress, encode, ffmpeg · ffmpeg.py · Media format conversion

Research Depth Levels

Not every request needs a full investigation. Choose the appropriate depth based on complexity, or ask the user if unclear. "Round" = one cycle of parallel tool calls. A single round can fire 3-5 searches/reads in parallel. Rounds are sequential — each round builds on what you learned in the previous one. Level · When to use · Minimum effort L1 — Quick lookup · Single fact, definition, simple question · 1 round, 1-2 searches

Step 1 — Define Scope

Before searching, clarify: What is the core question? What does the user actually need to know? What source types are needed — official docs, news, academic papers, user reviews? What dimensions matter (price, performance, compatibility, recency...)? How confident does the conclusion need to be?

Execute each dimension as a focused investigation cycle: For each dimension in the task list: Mark in-progress — Before starting a dimension, update its todo status to in-progress Search — Fire 1-3 parallel searches targeting this specific dimension Read — Open and read the most relevant sources from search results

Step 4 — Fact-Check

Do not skip this step. Cross-verify key data from multiple independent sources (aim for 2-3; more for L3) If only a single source exists, explicitly note: "Based on a single source — not independently verified" When sources contradict each other: Identify the reason (timing difference? different versions? conflicting interests?)

Fluxly profile Runtime and permissions are inferred by Fluxly static checks, not author declarations or runtime test results.
Domain
Data
Author-declared agents
No explicit declaration found; this is not inferred or tested compatibility
Static check
88 / 100 · heuristic scan, not runtime safety proof
Runtime evidence
Not run; no isolated runtime test evidence is currently recorded
Author / version / license
@BrikerMan · no license declared
Token usage
Moderate
Setup complexity
Guided setup
External API key
Not required
Operating systems
Unspecified (assume cross-platform)
Runtime requirements
No explicit requirements detected
Detected file/system behavior
  • Read-only
  • Write / modify
  • Shell exec
Network behavior
External requests
Install commands
None (reference only)

Heads up: Tool permissions are not declared; actual access depends on the runtime.

Output preview skilless-ai-research.preview
# Step 5 — Synthesize Output

- **Conclusion first** — Lead with the most important finding or recommendation, then expand
- **Cite every data point** with `[1]`, `[2]`, etc., so the reader can verify
- **Be honest about uncertainty** — If evidence is limited, say so; do not force a conclusion
- End with a full source list

Discussion

Powered by GitHub Discussions. Sign in with GitHub to comment, react, or subscribe.