agentskill

Name: agentskill
Author: KevinSpringer1

30stars0forks

How to install agentskill?

kevinspringer1/conserved-primer-design/agentskill

$npx -y skills add KevinSpringer1/conserved-primer-design --skill agentskill

Installs into the current project.

Summary

Use when an agent needs to call the VirusPrimerPro Agent API instead of the GUI, or prepare viral genome inputs for it: request/use API keys, download viral genomes from NCBI, install/run MAFFT, check and repair abnormal sequences, prepare phylogenetic trees/annotations, upload a

Files · 1

View on GitHub

SKILL.md

1	# VirusPrimerPro Agent API
2
3	Use this skill when a task requires calling VirusPrimerPro through `/api/v1/agent/*` rather than the browser GUI. The API is asynchronous: submit files and parameters, start the run, then poll status until `succeeded` or `failed`.
4
5	## Setup
6
7	Prefer environment variables so secrets do not appear in prompts, logs, or commits:
8
9	```bash
10	export VPP_AGENT_BASE_URL="http://127.0.0.1:3006"
11	export VPP_AGENT_API_KEY="vpp_live_<prefix>_<secret>"
12	```
13
14	For admin-only key issuance:
15
16	```bash
17	export VPP_AGENT_ADMIN_TOKEN="<admin token>"
18	```
19
20	## Preferred Helper
21
22	Use the bundled standard-library Python helper for reliable uploads and polling:
23
24	```bash
25	python agentskill/scripts/vpp_agent_api.py \
26	--base-url "$VPP_AGENT_BASE_URL" \
27	--api-key-env VPP_AGENT_API_KEY \
28	submit \
29	--msa aligned.fasta \
30	--profile standard \
31	--config-json '{"mode":"primer_site","target_coverage":0.95,"amplicon_min":100,"amplicon_max":400}' \
32	--wait \
33	--poll-interval 30 \
34	--output-dir results
35	```
36
37	Import the helper when building a larger agent workflow:
38
39	```python
40	from pathlib import Path
41	from agentskill.scripts.vpp_agent_api import VirusPrimerProClient
42
43	client = VirusPrimerProClient.from_env()
44	run = client.submit_primer_site_run(
45	msa_path=Path("aligned.fasta"),
46	config={"mode": "primer_site", "target_coverage": 0.95},
47	profile="standard",
48	wait=True,
49	poll_interval=30,
50	)
51	print(run["status"])
52	```
53
54	## Workflow
55
56	1. Ensure the agent has an approved API key. If not, call `apply-key` and wait for an administrator to approve and issue a key.
57	2. Create a run with top-level `mode`, `profile`, `config`, and file metadata.
58	3. Upload each file in chunks. The chunk upload path must include `/parts/{part_no}`.
59	4. Call `/start`. It returns quickly with `queued`; do not wait on the HTTP request.
60	5. Poll `GET /api/v1/agent/runs/{run_id}` every 30 seconds, or longer for heavy runs.
61	6. On `succeeded`, fetch `/result` and artifacts. On `failed`, report `process_exit_code`, `warning_flags`, and any available artifacts.
62
63	Status handling:
64
65	- `created` or `uploading`: finish missing uploads, then call `start`.
66	- `queued`: worker has not started; wait and poll again.
67	- `running`: computation is active; wait and poll again.
68	- `succeeded`: fetch result and artifacts.
69	- `failed`: surface the error metadata and available artifacts.
70	- `aborted`: the upload run was cancelled before queueing.
71
72	## PhyloGuide Preparation
73
74	For `phylo_guide`, prepare the tree and annotation before calling `submit`. Do not invent genotype groups from the sequence names. Use literature or curated references for the target species, then make those groups explicit in a `groups.json` file.
75
76	```bash
77	python agentskill/scripts/vpp_agent_api.py prepare-phylo \
78	--msa aligned.fasta \
79	--groups-file groups.json \
80	--species "Target species name" \
81	--reference-id NC_000000 \
82	--target-group GenotypeA \
83	--citation "Author Year DOI/PMID or URL" \
84	--out-dir phylo_ready
85	```
86
87	Then submit:
88
89	```bash
90	python agentskill/scripts/vpp_agent_api.py \
91	--base-url "$VPP_AGENT_BASE_URL" \
92	--api-key-env VPP_AGENT_API_KEY \
93	submit \
94	--mode phylo_guide \
95	--msa phylo_ready/aligned.normalized.fasta \
96	--tree phylo_ready/tree.newick \
97	--annotation-file phylo_ready/annotation.json \
98	--config-json '{"mode":"phylo_guide"}' \
99	--wait
100	```
101
102	Read `references/phylo_prepare.md` before preparing a PhyloGuide job.
103
104	## NCBI + MAFFT Preparation
105
106	Use `agentskill/scripts/ncbi_mafft.py` when the agent must download viral genomes from NCBI, install or run MAFFT, check alignment quality, or repair abnormal sequences.
107
108	```bash
109	python agentskill/scripts/ncbi_mafft.py install-mafft
110
111	python agentskill/scripts/ncbi_mafft.py download-ncbi \
112	--query '"Target virus"[Organism] AND "complete genome"[Title]' \
113	--max-records 50 \
114	--output ncbi_raw.fasta \
115	--manifest ncbi_manifest.json
116
117	python agentskill/scripts/ncbi_mafft.py pipeline \
118	--input ncbi_raw.fasta \
119	--out-dir mafft_ready \
120	--thread 4 \
121	--adjust-direction \
122	--drop-reason empty_after_removing_gaps \
123	--drop-reason high_n_fraction
124	```
125
126	Use `mafft_ready/aligned.repaired.fasta` as the MSA for `prepare-phylo` or `submit`. Read `references/ncbi_mafft.md` before running the pipeline.
127
128	## Safety Rules
129
130	- Never print, commit, or write the full API key unless the user explicitly asks to capture a newly issued key.
131	- Prefer `Authorization: Bearer ...`; `X-API-Key` is also accepted.
132	- Use top-level `profile`; the API treats it as authoritative and overwrites conflicting `config.primer_site_profile`.
133	- `high_coverage` and `di

People who liked this also viewed

6 picks

Type

agentskill

30stars0forks

View on GitHub

How to install agentskill?

kevinspringer1/conserved-primer-design/agentskill

$npx -y skills add KevinSpringer1/conserved-primer-design --skill agentskill

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# VirusPrimerPro Agent API
2
3	Use this skill when a task requires calling VirusPrimerPro through `/api/v1/agent/*` rather than the browser GUI. The API is asynchronous: submit files and parameters, start the run, then poll status until `succeeded` or `failed`.
4
5	## Setup
6
7	Prefer environment variables so secrets do not appear in prompts, logs, or commits:
8
9	```bash
10	export VPP_AGENT_BASE_URL="http://127.0.0.1:3006"
11	export VPP_AGENT_API_KEY="vpp_live_<prefix>_<secret>"
12	```
13
14	For admin-only key issuance:
15
16	```bash
17	export VPP_AGENT_ADMIN_TOKEN="<admin token>"
18	```
19
20	## Preferred Helper
21
22	Use the bundled standard-library Python helper for reliable uploads and polling:
23
24	```bash
25	python agentskill/scripts/vpp_agent_api.py \
26	--base-url "$VPP_AGENT_BASE_URL" \
27	--api-key-env VPP_AGENT_API_KEY \
28	submit \
29	--msa aligned.fasta \
30	--profile standard \
31	--config-json '{"mode":"primer_site","target_coverage":0.95,"amplicon_min":100,"amplicon_max":400}' \
32	--wait \
33	--poll-interval 30 \
34	--output-dir results
35	```
36
37	Import the helper when building a larger agent workflow:
38
39	```python
40	from pathlib import Path
41	from agentskill.scripts.vpp_agent_api import VirusPrimerProClient
42
43	client = VirusPrimerProClient.from_env()
44	run = client.submit_primer_site_run(
45	msa_path=Path("aligned.fasta"),
46	config={"mode": "primer_site", "target_coverage": 0.95},
47	profile="standard",
48	wait=True,
49	poll_interval=30,
50	)
51	print(run["status"])
52	```
53
54	## Workflow
55
56	1. Ensure the agent has an approved API key. If not, call `apply-key` and wait for an administrator to approve and issue a key.
57	2. Create a run with top-level `mode`, `profile`, `config`, and file metadata.
58	3. Upload each file in chunks. The chunk upload path must include `/parts/{part_no}`.
59	4. Call `/start`. It returns quickly with `queued`; do not wait on the HTTP request.
60	5. Poll `GET /api/v1/agent/runs/{run_id}` every 30 seconds, or longer for heavy runs.
61	6. On `succeeded`, fetch `/result` and artifacts. On `failed`, report `process_exit_code`, `warning_flags`, and any available artifacts.
62
63	Status handling:
64
65	- `created` or `uploading`: finish missing uploads, then call `start`.
66	- `queued`: worker has not started; wait and poll again.
67	- `running`: computation is active; wait and poll again.
68	- `succeeded`: fetch result and artifacts.
69	- `failed`: surface the error metadata and available artifacts.
70	- `aborted`: the upload run was cancelled before queueing.
71
72	## PhyloGuide Preparation
73
74	For `phylo_guide`, prepare the tree and annotation before calling `submit`. Do not invent genotype groups from the sequence names. Use literature or curated references for the target species, then make those groups explicit in a `groups.json` file.
75
76	```bash
77	python agentskill/scripts/vpp_agent_api.py prepare-phylo \
78	--msa aligned.fasta \
79	--groups-file groups.json \
80	--species "Target species name" \
81	--reference-id NC_000000 \
82	--target-group GenotypeA \
83	--citation "Author Year DOI/PMID or URL" \
84	--out-dir phylo_ready
85	```
86
87	Then submit:
88
89	```bash
90	python agentskill/scripts/vpp_agent_api.py \
91	--base-url "$VPP_AGENT_BASE_URL" \
92	--api-key-env VPP_AGENT_API_KEY \
93	submit \
94	--mode phylo_guide \
95	--msa phylo_ready/aligned.normalized.fasta \
96	--tree phylo_ready/tree.newick \
97	--annotation-file phylo_ready/annotation.json \
98	--config-json '{"mode":"phylo_guide"}' \
99	--wait
100	```
101
102	Read `references/phylo_prepare.md` before preparing a PhyloGuide job.
103
104	## NCBI + MAFFT Preparation
105
106	Use `agentskill/scripts/ncbi_mafft.py` when the agent must download viral genomes from NCBI, install or run MAFFT, check alignment quality, or repair abnormal sequences.
107
108	```bash
109	python agentskill/scripts/ncbi_mafft.py install-mafft
110
111	python agentskill/scripts/ncbi_mafft.py download-ncbi \
112	--query '"Target virus"[Organism] AND "complete genome"[Title]' \
113	--max-records 50 \
114	--output ncbi_raw.fasta \
115	--manifest ncbi_manifest.json
116
117	python agentskill/scripts/ncbi_mafft.py pipeline \
118	--input ncbi_raw.fasta \
119	--out-dir mafft_ready \
120	--thread 4 \
121	--adjust-direction \
122	--drop-reason empty_after_removing_gaps \
123	--drop-reason high_n_fraction
124	```
125
126	Use `mafft_ready/aligned.repaired.fasta` as the MSA for `prepare-phylo` or `submit`. Read `references/ncbi_mafft.md` before running the pipeline.
127
128	## Safety Rules
129
130	- Never print, commit, or write the full API key unless the user explicitly asks to capture a newly issued key.
131	- Prefer `Authorization: Bearer ...`; `X-API-Key` is also accepted.
132	- Use top-level `profile`; the API treats it as authoritative and overwrites conflicting `config.primer_site_profile`.
133	- `high_coverage` and `di

People who liked this also viewed

6 picks

Type