agentsclimarketplace

Competitive landscape

Skill pivyme/suiperpower/core/skills/idea/competitive-landscape

Map competitors for a Sui idea, on Sui and adjacent chains. Use when the user wants a competitive landscape, competitor analysis, or to find their edge.From its SKILL.md

Install
npx -y skills add pivyme/suiperpower --skill competitive-landscape

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 12 stars12 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

6.4 KB, ~1.5k tokens by cl100k_base, as published. Nobody here has run it

Preamble (run first)

# Suiperpower telemetry. Routes through the CLI so projects.json + Convex stay in sync.
# Silent on failure so the skill never blocks. Opt out: SUIPERPOWER_TELEMETRY=off.
#
# AGENT NOTE: when this skill finishes, run the matching completion command:
#   suiperpower track competitive-landscape idea completed
# Or use "failed" / "aborted" if it ended that way.
command -v suiperpower >/dev/null 2>&1 && suiperpower track competitive-landscape idea started >/dev/null 2>&1 &
true

If TEL_PROMPTED is no, before doing real work, ask the user:

Help suiperpower get better. We track which skills get used and how long they take. No code, no file paths, no PII. Change anytime in ~/.suiperpower/config.json.

A) Sure, anonymous B) No thanks

Write the answer to ~/.suiperpower/config.json telemetryTier field and create ~/.suiperpower/.telemetry-prompted. Then continue.

What this skill does

Builds a structured competitive map for a candidate idea. Three rings: Sui-native competitors, adjacent-chain competitors, off-chain competitors (web2 services, centralized products). For each ring, lists the players, their stage, what they do well, what they miss, and the candidate's angle against them.

Output is a defensible angle (one sentence) backed by the map. Without a defensible angle, the candidate is "me too" and the skill says so.

When to use it

  • Mid-validation: the user has an idea and wants to see what already exists.
  • Pre-pitch: the user has a draft pitch and wants to substantiate "we are different from X" claims.
  • Post-launch: a competitor surfaced and the user wants to re-establish positioning.

When NOT to use it

  • The user has not picked an idea. Route to find-next-sui-idea.
  • The user wants validation overall, not just competition. Route to validate-idea (which calls this skill internally).
  • The user wants a specific protocol's research (DeepBook market gaps, Walrus storage gaps). Route to deepbook-research or walrus-research.

If you activated this and the user actually wants something else, consult skills/SKILL_ROUTER.md and hand off.

Inputs

  • .suiperpower/idea-context.md if it exists.
  • The category one-liner (e.g. "Sui-native lending", "consumer mobile zkLogin app", "DeepBook bot").
  • Optional: a list of competitors the user already knows about.

Outputs

A competitive block appended to .suiperpower/idea-context.md:

### Competitive landscape, <timestamp>

#### On Sui
- <project>: <stage>, <one-line summary>, <strength>, <gap>
- ...

#### On EVM / Solana / other chains
- <project>: <chain>, <one-line summary>, <why they have not come to Sui>
- ...

#### Off-chain
- <product>: <what they do>, <where they fall short for crypto-native users>
- ...

#### Defensible angle
- <one sentence stating the candidate's specific differentiation, naming at least one competitor it explicitly beats and how>

#### Risks from competition
- <if any incumbent could trivially extend into the candidate's territory, name them and the response>

Workflow

  1. Define the category

    • One sentence: what does the candidate do, and for whom?
    • Identify the closest comparable category (e.g. "lending" -> Compound, Aave on EVM; Scallop, NAVX on Sui).
  2. Map the on-Sui ring

    • Start with cli/data/clonable-repos.json and the broader Sui ecosystem catalog (when available).
    • Visit Sui-specific aggregators and explorers (SuiVision, Sui Foundation showcase, Web3Goyala, Suipiens).
    • List 3-7 named projects. For each: stage (idea, testnet, mainnet, dead), one-line summary, strength, observable gap.
  3. Map the adjacent-chain ring

    • Same approach for EVM and Solana incumbents.
    • For each adjacent-chain incumbent, ask: have they expressed interest in Sui? Why have they not migrated yet? (Often: liquidity, audience, cost of porting code).
  4. Map the off-chain ring

    • Centralized exchanges, web2 SaaS, traditional finance products that solve adjacent problems.
    • For each: what do they offer that is hard to match on-chain? What do they fail at that crypto-native users care about?
  5. Identify the defensible angle

    • One sentence. Names a specific competitor and a specific way the candidate beats them.
    • Tests: would a sophisticated Sui community member nod, or laugh? If it does not pass the laugh test, sharpen.
  6. Identify the risk

    • Could an existing on-Sui incumbent extend into this space in a single sprint? If yes, the moat is timing or distribution, not technology. Say so.
    • Could an adjacent-chain incumbent port to Sui in a sponsor-funded weekend? If yes, the moat must be Sui-native, not "first to Sui".
  7. Writeback

    • Append to .suiperpower/idea-context.md.

Quality gate (anti-slop)

Before reporting done:

  • Is each on-Sui competitor named with a real project, not a placeholder ("various lending dapps")?
  • Does the defensible angle name a specific competitor and a specific advantage, not generic phrases ("better UX", "faster")?
  • Is at least one off-chain comparator listed (most crypto products are competing with web2 alternatives more than they realize)?
  • Did the analysis honestly answer "could the incumbent trivially extend into this space"? If the answer is "we are not sure", that is a yellow flag worth recording.
  • Did the writeback happen?

If any answer is no, the skill keeps working.

References

On-demand references (load when relevant to the user's question):

  • references/where-to-look.md: Sources for Sui ecosystem mapping, adjacent-chain mapping, off-chain mapping.
  • references/defensibility-tests.md: Tests to apply to the candidate's angle.

Knowledge docs:

  • skills/data/sui-knowledge/06-opensource-research.md: Open-source research workflow.
  • cli/data/clonable-repos.json: Sui ecosystem repos catalog.

Use in your agent

  • Claude Code: claude "/suiper:competitive-landscape <your message>"
  • Codex: codex "/competitive-landscape <your message>"
  • Grok Build: run grok, then /competitive-landscape <your message> in the session
  • Cursor: paste a chat message that includes a phrase like "competitive landscape", or load ~/.cursor/rules/competitive-landscape.mdc and reference it.

If you activated this and the user actually wants something else, consult skills/SKILL_ROUTER.md and hand off.

What ships with it: 3 files

6.7 KB alongside SKILL.md

agents/

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.