AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Truefoundry Prompts

skill-truefoundry-skills-prompts · by truefoundry

Manages TrueFoundry Prompt Registry. Covers listing, creating, updating, versioning, tagging, and using prompts through the AI Gateway. Agent authoring belongs to the agents skill.

No reviews yet
0 installs
26 views
0.0% view→install

Install

$ agentstack add skill-truefoundry-skills-prompts

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-truefoundry-skills-prompts)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
4mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Truefoundry Prompts? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Prompts

Manage prompts in the TrueFoundry Prompt Registry.

Use this skill when the user wants to:

  • List prompts.
  • Inspect prompt versions.
  • Create a prompt.
  • Update a prompt by creating a new version.
  • Tag a prompt version, for example production.
  • Get a prompt version FQN for Gateway usage.
  • Generate code snippets or usage guidance for a prompt.

Do not create or edit agents here. Use truefoundry-agents for Agent Registry UI workflows.

Preflight

  1. Verify tfy login is complete. If not, use truefoundry-onboard.
  2. Confirm the target tenant.
  3. Confirm the ML repository or prompt repository where the prompt should live.
  4. For destructive actions, use dashboard-only guidance.

UI Path

The predictable UI path is:

TrueFoundry Dashboard -> AI Gateway -> Prompt Management / Prompt Registry

Use the UI when the user wants to visually create, compare, or test prompts.

Capability Matrix

| User intent | Supported path | Notes | |-------------|----------------|-------| | List prompts | API or UI | Present prompt name, ID, latest version, and tags. | | Inspect prompt versions | API, SDK, or UI | Include version ID/FQN and tags. | | Create a prompt | SDK or UI | Review messages/settings before creation. | | Update a prompt | SDK or UI | Treat as a new version; do not overwrite silently. | | Tag a version | SDK or UI | Confirm before moving stable tags like production. | | Get prompt FQN | SDK or UI | Use for Gateway and Agent references. |

List Prompts

API fallback:

TFY_API_SH=~/.claude/skills/truefoundry-prompts/scripts/tfy-api.sh
$TFY_API_SH GET /api/ml/v1/prompts

Present:

Prompts
| Name | ID | Versions | Latest | Tags |

Create or Update Prompt

Creating a new prompt or adding a new version should be treated as a reviewed change.

Collect:

  • Prompt name
  • Repository / ML repo
  • System and user messages
  • Variables
  • Model, if the prompt stores one
  • Temperature / top-p / max tokens, if needed

Show the final prompt content and settings before creating or updating.

Review format before create/update:

Prompt Change
| Field | Value |
|-------|-------|
| Name | my-prompt |
| Repository | ml-repo-fqn |
| Change | create prompt / create new version |
| Variables | user_input |
| Model | model-catalog:openai:gpt-4 |
| Tags | none / production |

Then show the messages in order:

System:
...

User:
...

SDK shape:

from truefoundry.ml import ChatPromptManifest

client.prompts.create_or_update(
    manifest=ChatPromptManifest(
        name="my-prompt",
        ml_repo="ml-repo-fqn",
        messages=[
            {"role": "system", "content": "You are a helpful assistant."},
            {"role": "user", "content": "{{user_input}}"},
        ],
        model_fqn="model-catalog:openai:gpt-4",
        temperature=0.7,
        max_tokens=1024,
        top_p=1.0,
    )
)

Versioning and Tags

Prompt edits create versions. Use tags such as production or staging to provide stable references.

Before applying a tag, show the current tag target if known and ask for explicit confirmation when the tag is stable or production-facing.

SDK shape:

client.prompt_versions.apply_tags(
    prompt_version_id="version-id",
    tags=["production"],
    force=True,
)

Get Prompt Version FQN

Use the UI or SDK to retrieve the FQN:

client.prompt_versions.get_by_fqn(fqn="ml-repo:prompt-name:production")

The FQN is used by Gateway prompt run flows and by TrueFoundry Agents.

Delete

Do not delete prompts or prompt versions from the agent. If deletion is requested, direct the user to the Prompt Registry UI.

  • The user can list and inspect prompts.
  • Prompt create/update is reviewed before execution.
  • Prompt version tags are explained and applied only with confirmation.
  • Prompt version FQNs are surfaced clearly.
  • Agent authoring is routed to truefoundry-agents.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.