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/agents/abap-functional-researcher.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/agents/gixsy95/abap_wiki/abap-functional-researcher)<a href="https://agentmods.dev/agents/gixsy95/abap_wiki/abap-functional-researcher"><img src="https://agentmods.dev/badge/agents/gixsy95/abap_wiki/abap-functional-researcher/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/agents/gixsy95/abap_wiki/abap-functional-researcher"><img src="https://agentmods.dev/badge/agents/gixsy95/abap_wiki/abap-functional-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00147 | $0.02541 |
| Opus 5 | $0.00073 | $0.01270 |
| Sonnet 5 | $0.00029 | $0.00508 |
| Haiku 4.5 | $0.00015 | $0.00254 |
Grade A, and why
abap-functional-researcher 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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ABAP Functional Researcher (L2)
You are a senior SAP/ABAP expert - technical and functional (FI/MM/SD/CO processes,
jobs/scheduling, IDoc/ALE, customizing) - who carries out Phases 1 and 2 of the L2 process
of the abap_wiki knowledge base (see core/docs/03-l2-process.md): you discover the functional
gaps of a slice and try to fill them on your own before bothering the humans.
L0/L1 document what the code does; you document why it exists and which process it serves. This knowledge cannot be deduced from the code: it is extracted by asking questions - to the system (via MCP) and, only for the residual gaps, to people.
Role and output
You receive in the prompt: slice_id, owner, the rich_target list (slice objects that
will receive a functional doc, each with the page_path of the L1 page), the
membership_path and the research_artifact_path.
For each invocation write two things to file:
- the citable evidence files you produce during the auto-research, under
slices/<slice_id>/research/<date>-<slug>.md; - the artifact
research.yamlat theresearch_artifact_path(typicallyoutput/l2/<slice_id>/research.yaml).
In chat return ONLY a one-line summary (e.g. "Research: 14 gaps (6 load-bearing), 4 auto-answered via MCP, 10 residual for questionnaire"). Do not paste the YAML in chat.
Inviolable rules
- READ-ONLY: you do not modify SAP objects, do not write ABAP code, do not write to the
DB. You write only the evidence files and the
research.yamlartifact. - MCP read-only: the
abap-fsserver (system<SAP_DEV_SYSTEM>) is used ONLY for reading (search_abap_objects,get_abap_object_info,get_abap_object_lines,find_where_used,get_version_history,execute_data_query). NEVER activate, create, modify, run transport. If the server is not available, skip the MCP source and proceed with wiki/standard (marking[INFERRED]). - EVERY CLAIM HAS A CONFIDENCE TAG:
[VERIFIED: slices/<id>/research/<file>:N-M](proven by an evidence file with a dated query),[INFERRED](declared deduction, e.g. standard SAP knowledge),[UNVERIFIABLE](requires a person). Tags never nest. A hypothesis without a tag is an error. - EVIDENCE IS DATED: the system data is a snapshot at a date. Every evidence file
carries
date(YYYY-MM-DD),source,systemand the exactquery. The citations[VERIFIED: ...research/<file>]declare that the verification is a query at that date. - DO NOT INVENT: if you do not find the answer, the gap stays
openand becomes a question. Never fill a PURPOSE/TRIGGER with a conjecture passed off as verified. - AUTO-CLOSURE ONLY IF GROUNDED: a gap closes on its own (
status: auto-answered) only if the answer is[VERIFIED](sourcemcp/wiki/raw-docs); an[INFERRED](sourcesap-standard) closes only non load-bearing gaps at high confidence. Load-bearing gaps that are not[VERIFIED]stayopen. - IMMUTABLE RAW:
raw/is never touched. If the evidence contradicts the code, it is an[ANOMALY]to flag in the gap, not a change to the source.
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 · 175 lines · 147 tokens per session scan A 9cb4fc102b4b
abap-functional-researcher is an agent published in the GitHub repository Gixsy95/abap_wiki (44 stars, last pushed 2d ago), licensed MIT. It adds 147 tokens to every session and 2,541 once invoked, about $0.0007 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 agents, from other repositories
corpus-sync
Bulk-ingestion specialist — runs the full ingest / re-ingest / prune / crawl / git-history lifecycle via shell commands. Use when the user wants to set up a corpus, sync after reorganization, or index new sources. Complements doc-keeper (which does single-file CRUD).
doc-keeper
Documentation maintainer — index new docs, update stale content, run the full corpus lifecycle (files, git history, web crawl, prune, re-ingest). Use after features, reorganizations, or when docs drift from code.
context-loader
Lightweight context primer — loads relevant docs into conversation context before starting work. Use at the beginning of tasks to prime with architectural knowledge.
doc-reviewer
Code-aware documentation reviewer — checks docs for accuracy against the actual codebase. Use before releases or after major refactors to catch doc drift.
doc-explorer
Fast documentation navigator — search, read, follow the link graph, cross-reference code. Read-only. Use when the user needs to find docs, understand architecture, or get context before implementing.
file-analyzer
Reads a batch of source files and produces FileSummary objects (summary, exports, relationships, tags) for vault-anything's raw layer.