Pharma DB Tools

What it does

When your agent uses it

• The user asks to look up compounds
• Measured binding affinities
• Regulatory labels
• Clinical trials

Example prompts

• “/pharma-db-tools”

Requirements

• Python 3

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

1. 1Save both a machine-readable result file and a summary JSON.
2. 2Report the exact database, mode, identifier, filters, and endpoint used.
3. 3Prefer exact identifiers when available: PubChem CID, ChEMBL id, NCT id, DOI.
4. 4Return direct stable links when the upstream database exposes them.
5. 5Distinguish clearly between compound metadata, activity measurements, regulatory evidence, clinical-study records, and literature hits.
6. 6If the API returns no hits, say that explicitly instead of inferring a scientific conclusion.
7. 7Treat these sources as evidence surfaces for prioritization and review, not as experimental proof.

What it can do on your machine

Read from SKILL.md and the folder at commit 960a6e0. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships script files (Python), which the agent can run.

  Shell commands in SKILL.md call:

  • python3

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Pharma DB Tools loads about 1.4k tokens when it runs. Until then it costs about 89 tokens; SKILL.md has 345 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~89

When it runs · the whole SKILL.md, loaded when a task matches

~1.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from DrugClaw/DrugClaw at commit 960a6e0, republished under its Apache-2.0 licence (© DrugClaw). 345 words, ~1,427 tokens.

name

pharma-db-tools

description

Query public drug-discovery and translational-research databases including PubChem, ChEMBL, BindingDB, openFDA, ClinicalTrials.gov, and OpenAlex. Use when the user asks to look up compounds, measured binding affinities, regulatory labels or adverse events, clinical trials, or drug-discovery literature from public APIs and curated exports.

source

drugclaw

updated_at

2026-03-11

which python3 || true
python3 - <<'PY'
mods = ["requests"]
for name in mods:
    try:
        __import__(name)
        print(f"{name}: ok")
    except Exception as exc:
        print(f"{name}: missing ({exc})")
PY

python3 templates/pharma_db_lookup.py pubchem \
  --query imatinib \
  --output pharma/pubchem_imatinib.csv \
  --summary pharma/pubchem_imatinib.json

python3 templates/pharma_db_lookup.py chembl \
  --mode molecule \
  --chembl-id CHEMBL941 \
  --output pharma/chembl_imatinib.csv \
  --summary pharma/chembl_imatinib.json

python3 templates/pharma_db_lookup.py bindingdb \
  --tsv BindingDB_All.tsv \
  --uniprot-id P00519 \
  --affinity-type Ki \
  --max-nm 1000 \
  --output pharma/bindingdb_abl1.csv \
  --summary pharma/bindingdb_abl1.json

python3 templates/pharma_db_lookup.py openfda \
  --endpoint label \
  --query imatinib \
  --output pharma/fda_imatinib_label.csv \
  --summary pharma/fda_imatinib_label.json

python3 templates/pharma_db_lookup.py clinicaltrials \
  --condition "non-small cell lung cancer" \
  --intervention osimertinib \
  --status RECRUITING \
  --output pharma/osimertinib_trials.csv \
  --summary pharma/osimertinib_trials.json

python3 templates/pharma_db_lookup.py openalex \
  --query "KRAS G12C inhibitor resistance" \
  --limit 20 \
  --output pharma/openalex_kras_g12c.csv \
  --summary pharma/openalex_kras_g12c.json

python3 templates/pharma_db_lookup.py pubchem \
  --cid 5291 \
  --output pharma/pubchem_5291.csv \
  --summary pharma/pubchem_5291.json

python3 templates/pharma_db_lookup.py chembl \
  --mode activity \
  --target-id CHEMBL203 \
  --standard-type IC50 \
  --limit 25 \
  --output pharma/egfr_ic50.csv \
  --summary pharma/egfr_ic50.json

python3 templates/pharma_db_lookup.py bindingdb \
  --tsv BindingDB_All.tsv \
  --compound-name imatinib \
  --affinity-type Ki \
  --limit 25 \
  --output pharma/imatinib_bindingdb.csv \
  --summary pharma/imatinib_bindingdb.json

python3 templates/pharma_db_lookup.py openfda \
  --endpoint event \
  --query pembrolizumab \
  --output pharma/pembro_events.csv \
  --summary pharma/pembro_events.json

python3 templates/pharma_db_lookup.py clinicaltrials \
  --nct-id NCT04280705 \
  --output pharma/nct04280705.csv \
  --summary pharma/nct04280705.json

python3 templates/pharma_db_lookup.py openalex \
  --author "Jennifer Doudna" \
  --limit 20 \
  --output pharma/doudna_works.csv \
  --summary pharma/doudna_works.json

© DrugClaw, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/pharma/pharma-db-tools of DrugClaw/DrugClaw.

• SKILL.md
• templates/pharma_db_lookup.py

Open the folder on GitHubat commit 960a6e0