Borrowing it
Nothing to install: this file belongs to Gixsy95/abap_wiki. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Gixsy95/abap_wiki/main/.agents/skills/research-l2/SKILL.mdgit clone --depth 1 https://github.com/Gixsy95/abap_wikiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/gixsy95/abap_wiki/research-l2)<a href="https://agentmods.dev/skills/gixsy95/abap_wiki/research-l2"><img src="https://agentmods.dev/badge/skills/gixsy95/abap_wiki/research-l2/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gixsy95/abap_wiki/research-l2"><img src="https://agentmods.dev/badge/skills/gixsy95/abap_wiki/research-l2.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00115 | $0.01222 |
| Opus 5 | $0.00057 | $0.00611 |
| Sonnet 5 | $0.00023 | $0.00244 |
| Haiku 4.5 | $0.00012 | $0.00122 |
Grade A, and why
research-l2 scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research L2 - gap discovery, auto-research, questionnaires
Promotes knowledge from what the code does (L1) to why it exists and which process it
serves (L2), per slice. The cycle is driven by you (main agent): you invoke the sub-agent
abap-functional-researcher, the scripts handle the rest (persistence, views, questionnaires).
Architecture (see core/docs/03-l2-process.md):
- researcher = sub-agent
abap-functional-researcher: reads the L1 pages of the rich_target + the graph, discovers functional gaps, attempts to answer them autonomously (wiki ->raw/docs-> MCPabap-fs<SAP_DEV_SYSTEM>read-only -> standard[INFERRED]), writes citable evidence files and theresearch.yamlartifact. - submit-research persists gaps+evidence and closes auto-answered gaps as
[VERIFIED]. - questionnaire triages the residual load-bearing gaps by recipient, with the pre-filled hypothesis: the expert confirms/corrects instead of writing from scratch.
Prerequisite: slice-init already run (the slice has membership + real owner). The
MCP abap-fs server running reduces the number of human questions; if it is off, auto-research
proceeds on wiki+standards and nearly all gaps become questionnaire items (still useful).
Cycle
-
Get the rich_target (the objects to document, with their L1 page paths):
.venv/Scripts/python core/src/tools/pipeline.py slice-targets --slice <slice-id>Returns JSON
[{slug, sap_type, sap_name, devclass, page_path, hop, role}, ...]. -
Fan-out of the researcher (one per slice, or in batches if the rich_target is large - e.g. 8-10 objects per invocation). For each batch, one
Task(subagent_type="abap-functional-researcher", ...)passing in the prompt:slice_id,owner(from the manifest),rich_target= the subset of objects in the batch (slug + page_path),membership_path = slices/<slice-id>/membership.md,research_artifact_path = output/l2/<slice-id>/<batch>/research.yaml. The agent writes the evidence files underslices/<slice-id>/research/and the artifact.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 84 lines · 115 tokens per session scan A b311df362362
research-l2 is a skill published in the GitHub repository Gixsy95/abap_wiki (44 stars, last pushed 2d ago), licensed MIT. It adds 115 tokens to every session and 1,222 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
link-memory
Use after important user-approved decisions, when durable context should be proposed or reviewed, and for explicit Link memory lifecycle work: remember, recall, review, update, archive, restore, forget, or explain local memories through the CLI without requiring MCP.
link-retrieve
Use before answering work that may depend on user memory, project history, source-backed notes, or prior decisions; retrieve compact Link context through the CLI without loading the whole wiki or requiring MCP.
link-health
Use at the start of Link work when readiness is unclear, after installs or upgrades, and before repairs; verify health, inspect interrupted writes, back up, and repair generated indexes without MCP.
link-ingest
Use when raw files are present, source pages look stale, or a user asks to ingest notes into Link; refresh source-backed wiki pages, propose memories, and validate updates through the CLI without MCP.
ingest
Populate the gnosis-mcp knowledge base — from local files, git history, or a crawled website. Handles the full matrix of flags (--force, --prune, --wipe, --embed, --include-crawled) in one place.
manage
CRUD operations on the knowledge base — add, delete, update metadata. For bulk ingest / re-ingest / prune, use /gnosis:ingest instead. Requires GNOSISMCPWRITABLE=true.